Market Research Methods: Quantitative & Qualitative Insights

Let's suppose we are conducting research to find out the number of patients that would be visiting any particular hospital (let's take Indonesia for reference) for XYZ reasons, either for routine check-ups or any kind of treatment. So, I would like to determine factors like the number of patients that have visited that hospital for any particular time stamp, and then I would like to determine a series of factors like the number of beds, inpatients, outpatients, BOR (Bed Occupancy Rate), ALOS (Average Length of Stay) by each specific category (or split by specialty to be more precise). Can you suggest the illustrative data for the above case?

Friends, let us start this course with the opening case study that will illustrate to you the use of quantitative analysis in business situations, in the needs of investors or business people.

They are looking for data and information that will be very helpful for making investment decisions, projecting demand, and creating the infrastructure for upcoming demand.

How quantitative analysis is used in those situations.

This case study will tell you.

Let us look at this opening case study.

In this opening case study, we are talking about one investor who wished to invest in creating a multi-specialty hospital.

It involves making decisions regarding the different types of costs that are involved in the project.

Uh, what will the size of the demand be?

What should be the facilities?

Those are to be provided in the Multi-speciality hospital.

The personnel, the equipment, and the devices that are to be created for a projected demand.

He wants to optimize the initial infrastructure to start with and eventually learn from the data, the real data, and make some modifications.

What happens?

This investor is asking its market research team to give answers to some of the questions of the investor.

The first question the investor is asking the research team.

He says Let's suppose we are researching to find out the number of patients that would be visiting any particular hospital.

Let's take Indonesia, for example, because in this particular case study, this investor happens to be from Indonesia.

He is taking Indonesia as an example for whatever may be the reason x, y, z reasons either for routine checkups or any kind of treatment.

Maybe special treatment also.

So he says that I would like to determine factors like the number of patients that would be visiting or if they have already visited that hospital in the future, when the hospital is ready.

He is asking that he would like to know the number of patients who may be visiting that hospital at any particular time stamp.

When we say time stamp, we are saying every month or every quarter, or on an annual basis.

Then he says that I would like to determine a series of factors, like the number of beds in patients, the outpatient bed occupancy rate, the average length of stay of the patients by each specific category, or, in other words, split by specialty to be more precise.

He says Can you assist with the illustrative data for this particular Situation.

This is the first question that he is asking the research team.

Now the research team comes out with one illustrative data point in the form of a table that is based on the table provided by the investors.

The investor has provided this particular table that you can see here.

In this particular table, he has given, uh, the different columns and rows where he has explained his requirement according to the question that was asked.

He is asking for the number of beds, the inpatient bed occupancy rate, and the average length of stay at the hospital level by specialties, including, for example, surgeries.

He has given in this particular table, as you can see here, different specialties, including the overall numbers.

Also, the total figure.

He talks of the split by specialty in percentage terms.

He is asking for uh, you know, categories like general surgery or gynae or pediatrics or internal medicine, and so on.

He has given a complete list that he has been able to find out from his initial research, what the different specialties and specialties that are required in a multi-specialty hospital, and the column, he is asking the question that he had asked in a particular year.

He is asking for 2022 as an example, only the number of beds, number of inpatients, number of outpatients, bed occupancy rate, and average length of stay in that particular year.

And by the split of the specialty, it will be the different percentages of the overall figure.

That is what he is asking.

Based on this query.

The research team comes out with this table.

In this table, as you can see here, they have given complete information about the number of beds, inpatients, outpatients, and their split, which has been explained in this particular illustration. This is helpful to the investor because by looking at this table, he will get some initial idea of what he is trying to do.

Maybe this split is not validated.

The split that is given here is not accurate, but it will give him some starting point with this question, since he is asking only an illustrative data note; the verified data, or the data that has been collected from some hospitals, are on a payment basis.

Not like that.

He is simply asking for example, for illustrative data based on the secondary research team has prepared this illustrative data based on this query of the investor.

You can see this data here.

It is very much indicative of the future figures, quantitative figures that will be made accurate, that will be validated, that will be investigated, and made to look like real figures later on.

The investor is further interested in finding the peak demand for the services and asks the following question to the research team.

"Now, after finding out a relevant split of patients under the various heads required, I would like to determine the demographic divide of patients, and I have attached a table below for your reference.

Now, the second query of the investor was that he asked the research team, after sorting out this data, was the illustrative data was what he wanted from the research team.

He says that I would like to know for which particular period of the year there would be more visits from the patients, which means those months are to be identified where the number of patients coming will be very high.

Why is it important?

The capacity, maximum capacity that is to be created in the Multi-speciality hospital would be able to cater to those rush hours, rush days, rush weeks, or rush months.

So whatever the figure comes, the investor will be taking it as a benchmark that that much capacity has to be created.

That's why it is very, very important to understand which months will have more visits.

And eventually, he would like to know actually the numbers also, and accordingly, the investment will be made, the investment will be made for the peak season.

It is not obvious.

He further says that firstly, it is not possible to have the same amount of traffic flowing each day or month.

That is, he is very accurate.

It would be varying, obviously it will be varying.

He says that I would like to know what the factors are that would play a role in deciding that, which are the months, and why those months are there.

What are the factors behind the upsurge or the surge in patient visits?

It means those factors would give some indication of the upswing, the number increasing to the peak, and down downsurge number getting reduced because when numbers are getting reduced, the investor would like to find some methods to reduce costs if he already knows that, which are the months that are lean season, he would do something to reduce costs.

At the very start, he will design it like that.

He will try to make sure that he is not at a loss.

Therefore, he wants to know that.

What are the factors that would play a role in determining the visits for a particular time frame?

Now the research team comes out with some information based on the secondary research, and they identify certain variables that play important roles.

These are the independent variables.

The dependent variable is the number of patients visiting the hospital.

These independent variables that are identified by the research team include.

The first one is seasonality.

The research team says that seasonal patterns can significantly impact patients' visits to the hospitals.

And in this particular case, the Upcoming hospital.

Certain illnesses or health conditions may be more prevalent during specific seasons.

This is what the research team says about this particular variable: seasonality.

It also says that, for example, respiratory illnesses tend to increase during the winter months.

Understanding these seasonal variations in the patient's visits can help identify peak periods.

So this factor decides the peak period.

This factor tells you what is the peak season.

This factor tells you, if not completely, but it contributes to the actual figures of the patients coming in different months.

Then another independent variable that has been identified by the research team refers to public holidays and festivals.

Patient visits to the hospitals can be influenced by public holidays and festivals. During certain holidays or festivals, there may be an increase or decrease in patient traffic depending on the nature of the event and cultural practices.

For example, during the festive seasons, there might be an increase in accidents or injuries that will lead to a greater number of patients visiting the hospital, leading to higher emergency room visits.

At the same time, certain public holidays may result in fewer visits by patients, especially those patients who are not in a hurry or in emergencies.

Another independent variable identified by the research team is the weather conditions.

It says that weather conditions, such as extreme heat or cold spells or heavy rainfall, can impact patients' visits.

Weather-related health issues, such as heat strokes or accidents caused by slippery roads, can lead to fluctuations in hospital visits.

This is the contention and the reasoning that has been done by the research team for this particular independent variable, that is, the weather condition.

Now, these three variables have been identified by the research team that I just mentioned: seasonality, public holidays, and festivals.

Also, weather conditions and other variables have been identified by the research team.

They are also independent.

Variables include variables like epidemics or outbreaks.

The reasoning given by the research team is the occurrence of epidemics or outbreaks such as influenza, dengue fever, or COVID-19.

Recently, there was can significant impact on patient visits.

During such periods, there may be a surge in patients seeking medical care related to specific illnesses.

That is the reasoning given by the research team on this particular independent variable.

It also identifies another independent variable, which is the public awareness campaigns that are happening through the efforts of NGOs by the efforts of the government.

Now, these public health campaigns or awareness initiatives can influence patient visits.

For instance, campaigns promoting preventive health checkups or specific screenings may lead to increased visits during the campaign period.

That is very, very interesting.

Therefore, it may be sponsored by hospitals.

Such kinds of campaigns.

Another variable that has been identified, an independent variable that has been identified by the research team, refers to the health care policies.

Now what happens?

The reasoning given by the research team about this particular variable, and the factors that are connected to this particular variable, are described as changes in healthcare policies, such as the introduction of new healthcare schemes, insurance coverage, modifications, or government initiatives that can affect patients' visits.

These policies may influence patients' behavior or access to health care services.

These independent factors can contribute to the final number of patients who visit the hospitals.

Other factors that have been identified by the research teams are the factors that are demographic factors.

So the research team says that demographic factors, including population growth, age distribution, and migration patterns, can impact patient visits.

Changes in the local population, such as an aging population or an influx of migrants, can also affect healthcare utilization patterns, which means the number of patients visiting.

That is, the dependent variable.

And finally, the research team has identified that socio-economic factors also influence the number of patients visiting.

Therefore, it says socioeconomic factors such as income levels, employment rates, and access to healthcare facilities can also play an important role in determining the patient's visits.

Economic downturns or changes in employment rates may impact healthcare healthcare-seeking behavior of the people and directly affect the patient visits the numbers.

With this, the research team was able to convince the investor that these are the factors and, uh, any kind of quantitative analysis that has to be done further would include these independent variables, uh, some kind of linear regression or multiple regression.

Uh, work has to be done to determine the relationship between these independent variables and the dependent variable, that is, the number of visits.

This is the starting point for carrying out quantitative research in this area.

Therefore, the research team says that by analyzing, for example, the historical data available from different sources, those sources will be discussed in later courses in this program, which would be a part of the market estimation or understanding the market dynamics, or forecasting those things will be discussed in future courses of this particular program.

Course two and course three.

In this case, the research team says that analyzing the historical data and taking into account these factors that we discussed, and performing statistical analysis, particularly time series modeling in this particular case, can do the trick to understand the number of actual visits of the patients or approximate visits of the patients in the hospitals.

It says that by doing this, the investor can identify patterns and trends in patient visits during specific time frames during peak season.

During the lean period, those patients and trends can be extrapolated using time series modeling to forecast those things that will be discussed in the later courses.

The research team says that this analysis can provide insights into the factors influencing the upsurge or decrease in inpatient visits, and it can help hospitals, or this particular hospital that is being built by the investor.

It can help hospitals to plan and allocate resources accordingly.

What will be the resources?

What will be the cost of these resources?

How many resources have to be deployed?

It can be worked out through forecasting and market estimation.

Now the investor wants to know the demographic divide of the projected patients and asks the following questions.

"Now, after finding out a relevant split of patients under the various heads required, I would like to determine the demographic divide of patients, and I have attached a table below for your reference."

Now, the next question the investor to the research team is he says Now, after finding out a relevant split of patients under the various heads required, he says that I would like to determine the demographic divide of the patients in, for example, in this case, Indonesia.

He has attached a table of the type of information and the format.

He wants this information for the reference of the research team.

This is the table that he has given this table.

You can see here that it talks about the patient type and the percentage that he says.

That should add up to 100% obviously, because these are the patients coming from uh local for example, in this case from DKI Jakarta or Greater Jakarta, or Rest of Java, or rest of Indonesia.

He has categorized these regions that form the complete possibility of the patients coming from different parts of Indonesia.

He has also identified another category, which is the medical tourists.

Those tourists are international patients coming from outside Indonesia.

He says that this percentage, he wants to understand approximately how it has to be done, and how it will be done, which will be discussed in the later courses because that is part of estimation, market estimation, market dynamics, and forecasting that we will be discussing in course two.

However, the research team has given this illustrative data based on historical data and based on the methods that they used for market estimation and forecasting.

This method, I will not be able to explain to you in this particular course because it is beyond the scope of this particular course, but this is what they gave as the illustrative data based on the historical data.

The investor now asks the following question to the research team.

"And now, I would like to have a forecast for the coming 5 years for the same information mentioned above. Now, to arrive at a number, I would have considered some things to arrive at a particular number. So, what factors should I take into consideration to, you know, arrive at a number/situation, and why should I consider these factors? What backing or probability does it have to substantiate my data/findings?"

The next question that comes from the investor to the research team is like this, he says.

And now I would like to have a forecast for the coming five years for the same information talked about.

Now, he says, to arrive at a number, I would have considered some things to arrive at a particular number.

So he says that what are all those factors that he should take into consideration to arrive at a number or a situation, and why he should consider these factors.

He is asking this thing to the research team what backing or probability it has to substantiate the data or the findings.

That is the question he has raised.

Now the research team comes out with the answer to this query of the investor.

Like this.

They say that there are some key factors to consider.

What are these key factors, uh, to satisfy the question that has been asked by the investor?

The key factors are first, population growth, and then.

Demographics.

The team says that changes in the population, including population growth, age distribution, and migration patterns, can influence the number of patients from different regions.

Consider the projected population growth and demographic trends to understand how they may impact healthcare utilization.

Other factors that have been suggested by the research team are the economic factors, economic conditions, and income levels that can affect the healthcare-seeking behavior of the patients.

Changes in employment rates, income distribution, and economic development in each region can also influence the number of patients seeking medical care.

This is another factor.

The third factor that has been identified by the research team refers to health awareness and education.

Public health campaigns, awareness initiatives, and education programs can impact healthcare utilization.

Consider the impact of such initiatives on the likelihood of patients seeking medical care.

Therefore, these factors have to be understood from secondary sources in Indonesia.

What is happening?

Starting with the demographic factors that I talked about, it is possible to find out the illustrative percentages that have actually been shown by the research team.

Another factor that is to be considered by the investor in this particular query refers to seasonality and weather patterns.

Seasonal variation in weather conditions can affect patient visits, analyze historical patterns related to seasonality and weather conditions to identify any recurring trends and their potential impact on patient flow and, very importantly, historical patient data that is very, very important.

Therefore, the need is to analyze the historical patient data, including the trends, patterns, and seasonality, to understand the historical behavior of patient visits from different regions that have been mentioned by the investor.

Utilize statistical techniques and time series analysis to identify any underlying patterns or correlations.

It will be required for forecasting purposes.

Historical data will be required for that.

The research team further says that by considering these factors.

The investor can develop a comprehensive forecast that incorporates various influencing factors.

However, the team says it is important to note that forecasting involves uncertainty, and the accuracy of the forecast depends on the quality of data, the modeling techniques used, and the assumptions that are made, it says.

Regularly revisiting and updating the forecast based on new data and real-time observations can help refine the accuracy of the predictions over time.

In this case, predictions will be required because he is asking for the percentages, and using those percentages, it is possible to forecast by looking at the total number.

Those percentages can be used as per the table that has been asked for by the investor to assess the impact of the independent variable.

These independent variables are on the dependent variables.

Regression analysis is to be carried out.

That is part of the quantitative analysis techniques.

Now the investor is interested in knowing the projected methods by which the patients are likely to make payments and asks the following question.

"And, after determining the number of patients, their demographic divide, and all things determined above, I would like to know what mode they use to make payments or clear their dues. I would further like to have a split based on the demographics of the patients, do the modes of payment differ from region to region, and if it does or do not, how can I determine that, and how can I validate my information? So, again, I am attaching a table below for your reference."

The next question that is asked by the investor to the research team that has been hired by this investor it says that.

And after determining the number of patients, their demographic divide, and all things that are determined till now from his different queries, he says that I would like to know what mode they use to make payments or clear their dues.

Therefore, he says that I would like to have a split based on the demographics of the patients.

It further asks, do the modes of payment differ from region to region, or if it does or do not, how can I determine and how can I validate that information?

So again, he has attached a table, which table for the reference of the research team.

And it has some rows and columns.

As you can see here in rows.

They are cash public insurance, private insurance, corporate insurance, and others.

And in the columns, it is overall national and local patients and medical tourists.

This is the breakup he wants.

Based on this, the research team has come out with this answer and the table.

As you can see here, this is the table that has been given to the investor by the research team.

It says that to analyze the modes of payment and their potential differences based on patients' demographics and the regions, the team says the investor can collect data on the payment methods used by the patients when clearing their dues, and the example is given there it is based on secondary research, the data available in the public domain and through surveys and lot of quantitative research that is required collection of data that is required.

They have come out with some idea illustrative idea of how The split the table according to the table that is demanded by the investor.

They have come out with this information, as you can see here.

The research team further says that to determine the payment mode splits that have been requested by the investor, the investor can collect data from patients' records or the billing system if it is available.

Because it's a new hospital, it is not available.

If it is available from other sources of other hospitals, or by conducting surveys or interviews, if the data is not available.

To gather information on the payment methods used by the patients.

By analyzing the data to determine the percentage distribution of different payment modes within each demographic, group, and region, it is possible that if this data is analyzed, this percentage distribution can be, and this example that I have just shown you is based on that only.

It further says to the investor that to validate the findings, one can compare the payment mode split across different regions and demographics.

For this, one has to look for any notable variations or trends in the data.

Now, data has to be collected for different regions through primary or secondary research.

Additionally, the team says one can perform a statistical analysis, such as a chi-square test, to determine if there are statistically significant differences in this split of the payment mode preferences among different demographics or regions in Indonesia, for example.

It further advises that regularly updating and analyzing the data over time can help validate and refine these findings.

It is also beneficial to compare the findings with industry reports, studies, or benchmarks to gain further insights and ensure the robustness of the findings.

It is possible.

The investor further asks the following question to the research team.

"Now I would like to know how much personnel/expertise (or to be more precise, the type of expertise, let's say, Physicians, Nurses, Allied health, etc.,) has the hospital deployed for each particular specialty and based on the expertise deployed in a particular category/specialty, how much of it is allocated to OPD and IPD. After that, I would like to know how much revenue each department/specialty brings in and how much is the cost of running the operations of that particular department/specialty is. Again, I am attaching a table for you to have a better understanding."

The next question that has been raised by the investor to the research team is as this, he says.

Now, I would like to know how much personnel or expertise, whatever you say.

Or to be more precise, the type of expertise, let's say, example physicians, nurses, and allied health professionals that the hospital deploys for each particular specialty, and based on the expertise deployed in a particular category or specialty, how much of it is allocated to, for example, OPD outpatient department, and IPD inpatient department?

After that, he said that I would also like to know how much revenue each department's specialty brings in, and how much is the cost of running the operations of that particular department's specialty is.

Again, he has attached a table for the research team to have a better understanding of what he is looking for.

So this is the table that he has provided.

As you can see here, this particular table has rows that indicate the different specialties, and different common specialties that are there in a multi-specialty hospital, including general surgery, gynecology, pediatrics, and so on.

And he is asking for it in the columns.

Number of physicians, number of nurses, number of allied health, and percentage of the OPD.

What is the share of OPD?

What is the share of IPD?

The estimated revenue of the Department.

Different departments.

The estimated cost of different departments.

This information has to be collected by the research team.

He has also given at the bottom of this table, as you can see here, one note that says that the percentage split by specialty should add up to 100% for each column, except for the percentage of OPD and IPD in the column.

It is not possible to bring it to 100%.

However, in the Rose, IPD, and OPD percentages should add to 100%.

Obviously, for each row, that is, for each specialty.

Very good clarification.

He has given the kind of information he wants from the research team.

Now the research team has come out with these illustrative data for the investor.

I want to show you from this opening case study how the research team came out with those data.

What exact process is used?

I may not be able to explain this particular case study because, in courses two and three of this complete market research program, you will find different methods.

In this particular first course of the program, we are talking about qualitative and quantitative research and what is involved.

Therefore, we are giving you some information that is provided by the research team, not the information.

It will otherwise take too much time.

The research team has explained how they have come out with this particular table that is required by the investor.

It says that to determine the expertise, deployment, and personnel allocation, one can refer to hospital staffing records, job roles, and organization charts.

In this case, they are not available.

They have to find this data from other hospitals through surveys or primary and secondary research.

The allocation percentages can be based on the workload flow and the specific requirements of each department or specialty.

That is the method that has to be used based on the data that is available.

Indicative data.

The revenue and the cost figures can be obtained from the financial records of the competitors in this case.

Now it is a challenge to find this information about the competitors.

It is not easy, but it is possible.

Or it can be from accounting systems or cost analysis reports.

The revenue represents the income generated by each department or specialty, which may include patient fees and insurance incurred in running the operations of each department or specialty, such as salaries, equipment maintenance, and administrative costs.

It further says that it is important to note that the revenue and cost figures provided in the table are illustrative examples and may vary based on the specific context and scale of the hospital.

What is the scale of the hospital?

Regular financial analysis, budgeting, and cost management practices should be followed to ensure accurate and up-to-date information.

In this case, the research team has come up with these indicative, illustrative figures based on primary and secondary research that has been done.

The investor is now interested in knowing what the most important equipment is required and its costs.

"Now, taking into consideration the specialties that we have been talking or finding numbers about, how would I determine the list of top medical devices used to diagnose/a particular illness. And after that, what is the cost of that particular device (now the ranges of devices can differ more than 100 to 200% in some cases, then how should I arrive at a particular price that the hospital would have purchased that device for?"

The next question that is asked by the investor to the research team is as this.

He says that now, taking into consideration the specialties that we have been talking about or finding numbers about, how would I determine the list of the top medical devices that are to be used to diagnose that particular illness?

And after that, what is the cost of that particular device?

Price ranges of the devices can differ by more than 100% to 200% in some cases.

Then, how should he arrive at a particular price that the hospital would actually decide to purchase that device for?

So that is the question that has been asked by the investor to the research team.

He is looking for the answers.

Research answers.

The research team says that to determine the list of top medical devices used for diagnosing a particular illness in specific specialties, one can follow these steps.

This is what the system says.

The first step is to conduct research.

In this primary and secondary research has to be done what it says that a review of medical literature, industry publications, and online resources is conducted to identify the most commonly used medical devices for diagnosing the targeted illness in the relevant specialties.

It further says to look for consensus among experts and medical guidelines.

That will be the real figure because there will be a lot of variations in the figures in different publications.

Here, they are focusing more on secondary research, but primary research may also be required in this case.

In the second step, the team says to consult medical professionals.

Herein, they say that they engage with the healthcare practitioners, specialists, and department heads in the respective specialties to gather insights on the medical devices they use for diagnosis.

They can provide valuable inputs based on their experiences and expertise.

It is possible.

In the third step, the research team says to analyze hospital records.

This hospital is new.

Therefore, the hospital records of other entities are required in this case.

To do that, the team says one has to examine the hospital's equipment, inventory, and procurement records to identify the medical devices that have been purchased and are commonly in use.

This can provide a comprehensive list of devices currently employed in the hospital for diagnosing specific illnesses across different specialties.

These are the steps.

The team further says that once we have compiled the list of top medical devices, determining their cost can be challenging due to variations in pricing.

However, one can consider the following approaches.

The first approach is to contact suppliers and manufacturers.

It says to reach out to medical device suppliers and manufacturers to inquire about the cost range for the specific devices of interest.

They can provide pricing information or direct you to authorized distributors who can assist you further.

This is the advice that has been given, and the research team will be doing this only, but for future research, this has been suggested.

And in the second step, they say that you can also see quotations and request quotations from multiple suppliers or distributors to compare prices.

They can help you understand the market range for the devices that you are interested in.

Based on the list of devices, it is possible, and it says that you should consider average pricing, gather information and data from multiple sources, and calculate an average price for each medical device.

This can provide a reasonable estimate of the typical cost.

Also, another approach that has been suggested by the research team includes consulting the industry.

Reports refer to industry reports and publications that provide information on the average cost of medical devices that are in question.

These reports often include market trends, price ranges, and insights into device costs.

Approximate device cost.

It is possible to do that.

The team further says that it is important to note that medical device pricing can vary significantly based on factors such as brand, model features, specifications, and local market conditions.

Therefore, arriving at an exact price may be challenging.

These are very approximate prices, however.

By conducting thorough research, consulting professionals, and considering multiple sources, one can gain a better understanding of the approximate cost range for the medical devices used in diagnosing specific illnesses in different specialties and departments.

It is possible.

The investor finally asks the research team the following question

"And lastly, based on the number of patient visits of that particular hospital, how would I determine the quantities of diagnostic equipment possessed and used by the hospital? And, after determining the number (post clarification on validation), how would we find out the utilization rates of every particular diagnostic equipment?"

The last question of the investor is it says that Last, based on the number of patient visits of a particular hospital, how would I determine the quantities of diagnostic equipment possessed and used by the hospital?

This is the query that has been given by the investor to the research team, and he has given this table to give this answer to this query.

You can see this table.

It says that please state the number of diagnostic equipment that is available in your hospital using the list below.

This information can only be obtained through primary and secondary research through other hospitals.

Diagnostic equipment.

The list is given in the rows, different rows, and the other information that is required in different columns is mentioned here.

As you can see in this table.

For clarification purposes, he has given this.

The research team says that to determine the quantities of diagnostic equipment possessed and used by the hospital based on the number of patients visited, as well as to calculate the utilization rates of each diagnostic equipment, one can follow these steps.

The first step that can be followed refers to identifying the diagnostic equipment.

What is the diagnostic equipment?

Create a comprehensive list of the diagnostic equipment used in the hospitals for various specialties.

This can be done through experts by interviewing the experts and the medical practitioners.

This can include devices such as X-ray, machines, MRI scanners, ultrasound machines, or CT scanners.

These can be the different devices.

Secondly, the team says to gather data on patient visits.

It says that as already discussed.

Obtain accurate data on the number of patient visits to the hospital within a specific time frame, as mentioned earlier.

This information can be obtained from hospital records, electronic health records (EHRs, and so-called administrative data that we have discussed earlier.

Thirdly, it says to determine equipment requirements.

Analyze the workload and demand for diagnostic tests based on the number of patient visits that we have already done that exercise.

Consider the specific requirements of each specialty and the types of tests to be conducted.

Analysis will help you determine the quantity of equipment needed to accommodate the patient's volume in the peak season, as well as in the lean season that we discussed earlier.

Complimentary copy download from the resource section of the next lecture

Hi there!

I hope you are doing well and making great progress in this course.

I wanted to take this moment to congratulate you on your remarkable progress in this course.

Your dedication and commitment to learning have truly impressed me.

I have been following your journey closely, and I must say, I am delighted with the efforts you are putting in. As a token of appreciation for your hard work, I would like to offer you a complimentary copy of my recent book on a similar topic that you are learning in this course, which I believe will further strengthen your understanding and grasp of the subject.

You can download the PDF copy of this book from the resource section of this lecture.

This course is part of the VJ Export Import Mastery Courses Series, a collection of 25 different courses targeting the area of export management, designed to equip you with the knowledge and skills needed to excel in the field of export and international trade.

On my part, I am committed to helping you expand your learning journey by providing access to more similar courses in the series. But at the same time, on your part, I have a small request for you as well.

Your feedback is incredibly valuable in refining this course and ensuring it remains world-class and continually improved to its best version.

I kindly ask you to leave a rating for the course along with your honest feedback, if you have not yet done so. Your input will help me continue to improve and tailor the course to meet your needs and those of future learners.

Thank you once again for your dedication and enthusiasm.

Keep up the fantastic work you are doing, and remember, I am here to support you every step of the way.

Together, let's continue the journey of learning and growth.

"Embark on a data-driven journey with our comprehensive section on Quantitative Research. Discover the power of numbers as we dive into statistical analysis, research design, and hypothesis testing. Learn how to collect, interpret, and draw insights from quantitative data to make informed decisions. With practical examples and hands-on exercises, this section equips you with the essential tools to conduct rigorous and impactful research. Sharpen your analytical skills and become a proficient quantitative researcher through this engaging and informative course section."

In the next lecture, the instructor will discuss the overview of Module 1.

Welcome back to this lecture.

Now, let us discuss this module.

The first module that I talked about is concerning quantitative research, which is covered in this module.

Let me explain to you that this module covers what are the different areas.

The basic structure of this module refers to the fundamentals of quantitative research.

What are the different fundamentals, like key concepts, methodologies, and data analysis techniques?

These are the three main fundamentals that are covered in this particular module.

The first module of this course.

Let us see what are the key concepts.

What are the methodologies?

What are the data analysis techniques that are covered in this particular module?

Let us see that if we talk about the key concepts that are covered in this first module, I will be introducing you to the basic concepts of quantitative research, including things like sampling, survey design, and data collection.

These are the key concepts that are covered in this module and the methodologies that are covered in this particular module.

Different types of quantitative research methodologies are covered in this module, such as cross-sectional studies, longitudinal studies, and experimental studies.

These are the three most popular methodologies.

These three categories of methodologies are used in quantitative research that I am going to discuss and refer to.

In this module.

I will also be discussing the advantages and disadvantages of each of these methodologies, and I will also provide you with some examples of these different categories of methodologies.

Very popular categories.

And then finally in this course, I will also cover the different types of quantitative data analysis techniques, giving you examples, and what techniques like descriptive statistics.

Inferential statistics and regression analysis.

These are the different categories of techniques.

There are different techniques that I will be talking about.

But these are the categories of the techniques that I just mentioned to you.

I will also be giving you examples of how these techniques can be applied in market research to identify what is the purpose.

What is our aim of this quantitative research to identify the patterns and trends in consumer behavior? If we talk of business, although these techniques also be applied to non-business applications?

But I am talking of the marketing side, especially since my focus is on marketing.

I am talking about the patterns and trends in consumer behavior.

How do you identify that?

That is the purpose that will be covered in this module.

And finally, we will conclude this module by discussing some of the challenges and limitations of quantitative research.

That is how we will end this module.

Market research is an essential tool for businesses to understand their customers, competitors, and market dynamics. One important aspect of market research is quantitative research, which involves the collection and analysis of numerical data. This data can provide valuable insights into consumer behavior, preferences, and attitudes towards products and services.

In the next lecture, the instructor will introduce the basic concepts of quantitative research, including sampling, survey design, and data collection.

Friends, welcome to the next lecture in module one.

In module one, we will first be talking about the key concepts.

As I had explained to you.

What are the key concepts?

Three main key concepts are there.

First is the sampling in quantitative research.

Sampling refers to the process of selecting a subset of the population to participate in a research study.

How do you select the subset of the population?

That is a very professional work that requires a lot of skills.

A well-designed sample should be representative of the larger population.

How do you decide what kind of decision you take to estimate the sample you are taking?

You cannot take the whole population.

The sample that you are taking, how you are selecting that sample, and how the sampling is being done to make sure that it is nearly representative of the population, and you can avoid bias.

For example, suppose you are surveying school students.

Schools are of different categories.

They are government schools.

If we talk about the Indian context, there are government schools, there are private schools, and there are schools that are meant for the rural population of students.

Some schools are meant for the urban population of students.

How do you decide the student sample to be taken from which school, and what are the questions of your survey?

What is the survey design that we will be talking about?

Depending on the context, depending on the research outcomes that you are expecting or you are trying to test, what hypothesis you are trying to test, and depending on all, and depending on the focus groups.

Suppose your survey is focusing on the urban population.

Then that's a different context you are looking at.

The urban population means urban students in schools in urban areas.

But there are different categories of these schools in the urban areas also.

How do you do the sampling?

How do you select the subset of the population that will participate in a research study?

It requires a lot of skills and a lot of brainstorming.

The second concept of quantitative research refers to the survey design.

Survey design involves developing a set of questions that will be used to gather data from the sample population.

How you design those questions and what kind of method you are going to use will also decide the survey design.

Suppose, for example, you are collecting data through the internet through online methods.

Then, probably the survey design may be of a different type.

In another scenario, where you are collecting data on the street in the field, the survey design will be different.

Depending on the type of sample depending on the methods of data collection, which we will be talking about in the data collection part, the survey design will differ.

But the survey design has to focus on the outcome.

What hypothesis are you testing, and what is the context?

Survey design will involve developing the most suitable set of questions that is suitable for the context of the research.

The third key concept of quantitative research is data collection.

Data collection involves the administration of the survey.

The survey that you have designed for the sample population, the sample population that you have selected, the subset of the population that you have selected as a sample, and being able to effectively and accurately collect the responses.

What is the best method?

Again, it will depend on the context. Again, it will depend on the research questions. Again, it will depend on the nature of the research, the ecosystem where this research is being carried out, and who the stakeholders of the research are.

Many variables will help you decide the method of data collection, what kind of data collection it will be, and how you administer the survey.

What is the involvement of the survey in this data collection?

And the method of data collection will be according to that.

In quantitative research, there are different types of methodologies that can be used to collect and analyze data. These methodologies can be broadly classified into three categories: cross-sectional studies, longitudinal studies, and experimental studies.

Cross-sectional studies involve collecting data from a sample population at a single point in time. These studies are useful for understanding the current state of a population or group, and for comparing different groups or variables. For example, a cross-sectional study may be conducted to understand the preferences of different age groups for a particular product.

Now, the second part of module one that I am talking about, which we are referring to at present, refers to the methodology, and the methodology I had explained to you.

There are three categories.

The first category I want to talk about, to start with, refers to the cross-sectional studies.

What is the methodology that you use that you call the cross-sectional studies?

Your research studies, the quantitative research studies, and the quantitative analysis that you are trying to do refer to the methodology of cross-sectional studies. This refers to collecting data from a sample population at a single point in time, at the cross-section.

You are focusing on a particular moment, a particular cross-section of time.

What are the advantages?

The main advantages of this methodological approach, again, will depend on the context—whether the cross-sectional studies can give you the answer to your question, and whether it is suitable in the context.

The main advantage is that it is quick and easy to conduct.

And what it does is provide a snapshot of the current state.

The advantage is that because it focuses on the current state and provides a snapshot of what happens, you get the real-time data of that particular point in time, which is free of many biases or many of the challenges and difficulties that you may come across if you are doing the study over time.

This snapshot is good in that way.

This kind of cross-sectional study, depending again on the context, can be used to test hypotheses of the relationship among different variables at that time.

In fact, when we talk of hypothesis testing of the relationship among different variables, and when we are doing cross-sectional studies, many of the exogenous variables or extraneous variables may emerge.

If you do this study over time, that will not come.

This is the advantage.

The disadvantages of cross-sectional studies, to start with, are that they cannot show the changes over time.

Suppose the context requires you to study the variation of the variables or the relationship among variables over time.

It cannot be done, obviously, and it may not be representative of the entire population because you are doing the study at a particular moment, at a cross-section.

What happens?

What is available at that time, you take it or leave it.

What happens to the sample that you are doing?

I am not saying that it is the case all the time, but the chances are there because since the cross-section is there and at that time, it is very difficult to really find a sample that is representative of the entire population in many contexts.

I am not saying that in all contexts, but in many contexts, it can happen that it may not be representative of the entire population that you want to study.

This is a disadvantage.

And the third disadvantage of cross-sectional studies is that there is a possibility of biases creeping into the sample selection and data collection process, again because of the limitation of time. Since you are doing things in the current state, in a snapshot, it is possible in many situations that such kinds of biases may creep into the sample selection and the data collection process.

If we take the example of the cross-sectional studies, where they can be used.

Suppose you want to understand the current state of consumer satisfaction for a particular product.

Suppose you want to study that.

Maybe you want to study among groups also, like age groups, different age groups.

At a given point in time, the younger population, the older population, or the teenagers—different age groups, persons, or different clusters—you can do that.

That is not a problem.

You can have clustered samples or a set of samples that you can select at a particular moment of time, and you can compare.

You can analyze that particular snapshot in the current state to find out customer satisfaction among the different age groups, and their satisfaction with a particular product.

These are the types of situations where cross-sectional studies can be very useful, and they have certain advantages.

They also have certain disadvantages.

Longitudinal studies involve collecting data from the same sample population over time. These studies are useful for understanding how variables change over time, and for identifying patterns or trends. For example, a longitudinal study may be conducted to understand the changes in consumer preferences for a particular product over a period of several years.

The second category of approaches that are used in quantitative analysis and methodology.

We call it the methodology.

The second category of the methodology that is used is called longitudinal studies.

What is this longitudinal study?

The main feature of the longitudinal type of study is that it involves collecting data from the same sample of the population, whatever the sample is, whether representative of the population or not.

Ideally, it should be representative of a particular population that you are trying to study, but it uses the same sample over time.

That is very important.

It is not a cross-section.

It is a study over time.

You are trying to study the changes over time.

That is the main advantage of this category of studies.

You can show the changes over time.

One advantage.

The second advantage is that it can identify, very importantly, the trends and patterns in consumer behavior or attitudes.

For example, you want to find out the change in the behavior and attitude of consumers for a particular product over time.

Another advantage of this type of study, longitudinal studies, is that they can control for individual differences over time. That kind of control is possible.

How do you select your sample? How do you design your survey? How do you do the data collection?

Those techniques that you use there can help you to control for individual differences over time.

One of the disadvantages of this type of longitudinal study is that it can be very expensive and time-consuming because you are surveying over time.

I will just give you an example.

Suppose some researcher is trying to find out the deficiency of vitamin D among teenage students.

And he is trying to find out the impact of fortified foods. After consuming fortified foods, what types of changes happen in the vitamin D levels among the students?

First, he wants to find out what is the prevalence of vitamin D deficiency among the students.

And then he is trying to find out what impact there is over time of the periodic administering of fortified foods, or giving the fortified foods to the students over a period of time.

How does the vitamin D level change?

Suppose that kind of study is being done.

It has to be a longitudinal type of study, but it will be very expensive because the team has to be collected to do the job, the field research job, over a period of time, several times, maybe, and it will consume time.

It will require a lot of man-hours to carry out this kind of study.

Obviously, it will be expensive, and the attrition and dropouts can be very high and can affect the representativeness of the sample.

Suppose in the example that I was referring to, the researcher is trying to find out the impact of fortified foods on the students regarding the level of vitamin D in their bodies.

The sample may have to be changed over time because some of these students are not available next time you visit them.

Some students have left for their homes for some reason.

What happens?

The sample may not always remain the same. The number of students may be reduced, or some new students may have to be added in between.

Some kind of attrition, some kind of dropouts.

Those are very common in these kinds of studies.

This is one major disadvantage.

The third type of disadvantage that can be there in longitudinal studies refers to the confounding variables.

What are the confounding variables?

Those variables that you are not actually studying but come in.

Those variables affect the outcome.

For example, in our example, I was talking about the research on the impact of fortified foods on the vitamin D deficiency levels among the students.

Some of the students may be in a lot of sports activities, and they are outdoors, while some of the students are not. They always remain studying indoors only.

What happens when the level of vitamin D changes among the students who are actively participating in many sports events of the college where the research is being done?

Their involvement in outdoor activities impacts vitamin D levels, and that variable is a third variable that was not part of the research, but it is affecting the outcome.

These kinds of confounding variables can creep in, and the smart researchers will try to take care of such confounding variables.

Those have to be taken care of.

These kinds of disadvantages can be there in longitudinal studies because of the involvement of the period.

The time gap is there.

These kinds of variables creep in.

Another example of the context where longitudinal studies can be useful, for example, is to study and understand the changes in consumer behavior and preferences for a particular product over time.

If you are trying to understand how the preferences and behavior about a particular product are changing among consumers, these kinds of studies can be very, very useful.

Suppose you are introducing a hair health product in the market for females.

Depending on the time, depending on the different cross-sections of time, it is possible that the preferences and behavior of the consumers, that is, the females who are using the hair products, will change. You wish to introduce a new range of hair products over time because of the different factors that you are trying to understand. Over the period, what are the other variables that can impact those changes in consumer behavior and preferences?

What are those variables you want to understand?

What are those variables and what are the changes that are happening, how the preferences and behavior are changing with time, and whether those changes are natural, or are affected by a third variable, extraneous variable, or confounding variable?

These kinds of studies cannot be done in a cross-section.

They cannot be done at a particular moment.

You cannot have a snapshot of the sample.

It is not possible.

You need longitudinal studies in such kinds of cases.

Experimental studies involve manipulating one or more variables in a controlled environment to observe the effects on another variable. These studies are useful for understanding cause-and-effect relationships between variables. For example, an experimental study may be conducted to understand the impact of a new marketing campaign on consumer behavior.

The third popular category of methodologies that are used for quantitative research, especially for business purposes, refers to experimental studies.

That is the third category.

Now experimental studies.

This kind of methodology involves manipulating one or more variables in a controlled environment.

And why do you do that?

To observe the effects of another variable, some other variable, on a particular phenomenon.

What is the impact?

What is the effect of certain variables while you are manipulating one or more other variables in a controlled environment?

When you say controlled environment, when you say manipulation of one or more variables, you are doing a kind of experimentation.

What is the advantage of such kinds of experimental studies?

The methodology that is used in quantitative analysis is that it can show cause-and-effect relationships.

The effect is there because of this cause.

This is the input.

This is the output.

And because of this input, this is the output.

Cause-and-effect relationship between variables that are being studied.

The second advantage of experimental studies is that they can be used to test the effectiveness of interventions or treatments.

Some kind of intervention.

Some kind of treatment is done by marketers in the market for the consumers, for the customers.

How effective is the new intervention?

How effective is the new treatment?

For example, you come out with a new marketing campaign.

What is the effectiveness of a new marketing campaign?

It is a new way of passing on the message.

Marketing messages to consumers in a new way.

When you are trying to do this kind of study, you are doing an experimental study to find out the effectiveness of the new way of doing the marketing campaign and the different messages.

What messages work?

What messages do not work?

Those kinds of work can be done.

In this kind of controlled environment, where you are able to manipulate one or more variables, it is possible by design and by the ecosystem of experimental studies that you control the extraneous or confounding variables.

It is possible.

I am not saying that it can be done in all cases.

Extraneous variables are not being studied.

You can control it depending on how you design your survey.

How do you select your subset from the population as the sample, and how do you carry out the data collection in experimental studies, also?

The whole experiment, how it is designed, can give you the capacity to control some of the extraneous variables that may impact the outcome that you don't want to study because they are out of context.

The disadvantages of experimental studies are:

The first one is that it may not be representative of the entire population being studied.

Because you are doing experiments on a selective set, there is a strong possibility that the sample may not be representative because it is very, very difficult to smartly select the subset of the population to carry out experiments.

To do experiments, you have to convince the subjects to be part of the study.

That process itself is very challenging.

What happens is that you make compromises, and because of those compromises, the sample may be a compromise.

It may not be fully representative of the entire population.

Again depends on the situation.

Again depends on the context.

The second disadvantage of experimental studies is that they may not be practical or ethical in certain situations, especially when you are doing some medical-related experiments.

For example, if we talk of the genetic impact of mental diseases among the samples, among the subjects that are part of the research study.

You are doing some mental health-related studies, and you are getting some results, and sometimes you are sharing those results with the subjects.

You may be putting the subjects that are part of the research in a situation where they have a bad feeling about their mental health.

Sometimes it is not practical to do certain kinds of experimental studies, and sometimes it is also unethical.

The third disadvantage of an experimental study is that there may be biases in the sample, especially when you are selecting the subset.

Those biases can creep in during the data collection process.

Again, because of the challenges that are involved in preparing the subset sample, which is likely to be representative of the whole population.

You are making compromises.

Biases can creep in.

One of the examples that I just talked about a little bit, I also gave a reference, is the situation where experimental studies can be done.

For example, so-called A/B tests. A and B tests, for example, evaluate the effectiveness of different types of marketing messages that are aimed at driving customer engagement and sales of a new range of products.

You have a sample A, you have a sample B. You randomly assign one particular marketing message to sample A, and randomly assign another different marketing message to sample B, and you look over time at the impact of the different marketing messages in driving customer engagement and sales figures.

This is one business example where experimental studies can be done, for example, using A/B tests or many other tests that are possible.

There are several different types of quantitative data analysis techniques that researchers use to analyze and interpret numerical data. Here are some of the common techniques:

Descriptive statistics: Descriptive statistics are used to summarize and describe the basic features of a data set. This includes measures of central tendency (mean, median, mode) and measures of variability (standard deviation, range).

Inferential statistics: Inferential statistics are used to make inferences about a population based on a sample of data. This includes hypothesis testing, confidence intervals, and regression analysis.

Data mining: Data mining involves using software to search for patterns and relationships within a data set. This can involve techniques such as clustering, association rules, and decision trees.

Machine learning: Machine learning involves using algorithms to automatically learn patterns and relationships within a data set. This includes techniques such as linear regression, logistic regression, and decision trees.

Simulation: Simulation involves using mathematical models to generate data sets that can be used to test hypotheses and make predictions. This can involve techniques such as Monte Carlo simulations.

Time series analysis: Time series analysis involves analyzing data that is collected over time. This includes techniques such as trend analysis, seasonal decomposition, and ARIMA modeling.

The choice of data analysis technique depends on the nature of the research question, the type of data being analyzed, and the level of statistical sophistication required to address the question. Researchers should carefully consider the advantages and disadvantages of each technique before selecting the most appropriate approach for their study.

Okay, friends, now let us talk about the last part of this module, module one.

And we are talking about, as I had discussed earlier, the quantitative data analysis techniques.

We are now focusing on the techniques.

What are the different techniques?

Popular techniques.

There are many, many types of techniques, but the most popular and commonly used techniques for quantitative data analysis include techniques like descriptive statistics, inferential statistics, data mining, and many others.

Let us first talk about descriptive statistics.

Let us try to understand what descriptive statistics are.

Descriptive statistics are used to summarize and describe the basic features of a data set.

What are these basic features?

Things like measures of the central tendency of the data set.

A very basic, very initial understanding of quantitative data analysis.

The central tendency could be things like mean, median, or mode, or the measures of the variability of the responses in the data.

The data that you have collected? What are the responses, and what are the variabilities?

Things like standard deviation or range.

The idea is just to describe the phenomena from the responses.

Very simple things.

The second technique that is very commonly used in quantitative data analysis is called inferential statistics.

What is inferential statistics?

Inferential statistics are used to make inferences about a population based on a sample data set.

Again, we are targeting a particular population depending on the quality of the data set, depending on the quality of the sample, and how representative the sample is of the entire population.

We can make inferences about this population based on the sample only because we cannot collect data from the entire population.

From the sample, we can make some inferences, like hypothesis testing, confidence intervals, or regression analysis.

This mathematical work can be done in this kind of inferential statistics.

It is purely based on finding out the inferences.

The third technique in quantitative data analysis that is very popularly used is called data mining.

What is data mining?

Data mining involves using software to search for patterns and relationships within the data set, within the sample.

The software can help us identify the patterns and relationships among the variables, the target variables, and the variables that we are looking for in that particular data set.

We are trying to mine, we are trying to dig out such patterns and relationships.

That is why it is called data mining.

It used to be done manually at some time, but now computer software is used.

In this, techniques like clustering, association rules, and decision trees can be used to dig out the patterns and relationships within the data set.

In the fourth place, the most popular quantitative data analysis technique that is presently used is called machine learning.

What is machine learning?

Machine learning involves using algorithms to automatically learn patterns and relationships within a dataset.

Here we are trying to use the algorithms.

Whether machine learning can help us by using algorithms to automatically learn, it is a learning process of the patterns and relationships.

We are not digging it out.

We are trying to learn from the dataset using the algorithm.

These algorithms can also be tweaked.

They can be modified.

That is part of the learning.

We are trying to learn from the data.

That is why it is called machine learning.

But it is not done manually.

It is done through computer software using algorithms.

These include mathematical techniques like linear regression, logistic regression, and decision trees.

The difference between data mining and machine learning is that in this technique, we are trying to learn.

That is the idea.

It is the learning process.

We are not digging it out.

We are not mining the patterns and relationships.

That is the difference.

In the fifth place, a very popular quantitative data analysis technique that is used is called simulation.

What is simulation?

Simulation involves using mathematical models to generate data sets.

Here we are simulating using the mathematical models.

Based on past experiences, we have mathematical models.

We apply those mathematical models to the responses in the sample data set, and we generate new data sets that can be used to test the hypothesis.

The result is that we can make predictions.

Simulation is based on past experiences and past mathematical models on certain types of data sets on certain populations.

This can involve techniques such as Monte Carlo simulations.

One very good example of using the simulation process is the patient data in hospitals.

By using different mathematical models, it is possible to find out the possibilities of certain types of patients who may be readmitted to the hospital.

The benefit is that we can predict that these patients are likely to be readmitted in the future.

If anything can be done about that, some methods can be used to avoid such readmissions or reduce such readmissions.

Predictions are required.

Mathematical models are in place, and with some tweaking, it is possible to generate data sets from the raw data.

This is how it is done.

Monte Carlo simulation is the most common simulation that is used in hospitals in this type of example.

This simulation technique is very, very popular.

In the last place, the sixth place, I am talking about time series analysis.

What is time series analysis?

This involves analyzing data that is collected over a time series.

Over time.

Again, it is very similar to looking at the changes over time.

These kinds of techniques include things like trend analysis, seasonal decomposition, or something called ARIMA modeling (autoregressive integrated moving average modeling).

These are all in present times, helped by computer software, and a lot of information can be obtained from these techniques.

Different techniques are applied in different situations.

The skill lies in knowing which technique to use in what situations, and that requires a lot of knowledge and experience.

All these quantitative data analysis techniques are very commonly used, and they get you very good results, very good outputs, despite several challenges that are there in dealing with quantitative data analysis techniques.

In the next lecture, I will talk about these challenges that are there in quantitative data analysis.

Now, friends, before we go forward, let us look at some of the examples of the use of these different quantitative data analysis techniques that I just discussed in the last lecture.

Starting with descriptive statistics, let us take this example.

Suppose a researcher wants to understand the distribution of income in a particular area, a city, or a region.

He wants to understand this distribution of income.

What the researcher can do is use descriptive statistics to calculate the mean, median, and standard deviation of the income by taking a data set that is representative of that particular region. In that sample, by doing the data collection, he can find out the mean, median, and standard deviation of the income.

He can also create a histogram to visualize the distribution of the income.

The histogram will give a further description of the sample and the population.

Another thing that he can do with that particular data is identify if there are any outliers.

That will give him a lot of insights into the description of the population that he is trying to study, related to the distribution of income.

Maybe some outlier is there—somebody is a billionaire who is living there, or a group of billionaires is living there—that does not represent the mean, median, or mode of the population.

Those outliers, the existence of those outliers, will give further insights into the description of the population.

This can be one example.

Similarly, if we take the example of inferential statistics, the use of the inferential statistics technique in quantitative data analysis:

Suppose a company wants to test whether a new advertising campaign has been able to increase the sales of a particular product or range of new products.

The company can use inferential statistics to compare sales data from before and after the campaign. They can organize a campaign focusing on the advertising messages to the customers, and before and after, they can collect data.

They can also carry out a t-test that could be used to determine whether the difference in sales before and after is significant or not.

This kind of t-test can be used.

Now, let us look at one example of the use of data mining techniques in quantitative data analysis. In data mining, we can take this example:

Suppose a retailer wants to identify which products are frequently purchased together.

This means the customers buy something, and with that, generally and frequently, they also buy something else.

What they can do is use association rules to analyze the sales data. Whatever data they have, they can use it as a sample and identify the patterns of co-occurrence between different products.

Using the association rules, it is possible to identify the patterns.

This kind of information on the identification of the patterns could be used to optimize the retail store layout or to create targeted marketing strategies or campaigns.

This is how the data mining technique can be used.

Similarly, we take one example of the machine learning technique in quantitative data analysis.

In this example, suppose a hospital wants to predict which patients are at higher risk of future readmissions.

What the hospital can do is use logistic regression to develop a predictive model based on patient data.

They already have patient data in their computers.

This data can be, for example, age, medical history, and the type of medications given to the patient, or obtained through different consultancies.

This model could be used to identify patients who are at a higher risk of readmission.

This information will allow the hospital to provide targeted interventions to reduce future readmissions for the benefit of the patients.

This is possible.

Similarly, we can take one example of the simulation technique that is used in quantitative data analysis.

Suppose a government agency wants to estimate the impact of a new policy on the economy.

The government wants to understand this.

What they can do is use the Monte Carlo simulation that I discussed to model the potential effects of the policy on different economic variables.

What kind of variables can these be? The GDP and the unemployment rate in the country.

What is the potential effect of the policy?

This kind of model could be used in the Monte Carlo simulation.

This information will help the government agency to inform policy decisions and to identify potential risks and uncertainties.

This is how, using the Monte Carlo simulation, this particular phenomenon can be carried out by the government agency with respect to the impact of the new policy on the economy.

Now, the last technique that we discussed is time series analysis.

Let us take one example for that, also.

Suppose a researcher wants to understand the seasonal patterns in a particular type of crime—maybe some heinous crime, some crime that is more frequent in a particular country, region, or city.

This researcher wants to understand the seasonal patterns.

The researcher can use time series analysis. The data that is available to the crime agencies can be analyzed to identify whether there are any recurring patterns in the data over time.

If the researcher can identify any recurring patterns, this information could be used to allocate resources more effectively or to develop targeted intervention strategies to avoid such crimes in that particular area.

Those patterns can be helpful if they are identified by the researcher.

These were some of the examples that I wanted to discuss concerning the different techniques that I have talked about in this course, as far as quantitative data analysis is concerned.

In the next lecture, we will conclude this particular module, module one, by discussing the challenges that are there in quantitative data analysis.

Despite the wide use of quantitative research, it has several challenges and limitations. Dr. Jain discusses the challenges and limitations of this type of research in the next video. He also talks about the biases that can creep into research based on quantitative analysis.

The first very important challenge of quantitative analysis refers to its limited context.

Quantitative research often focuses on numerical data and may not be able to capture the broader context of a research topic.

It may not account for, for example, things like social, cultural, or historical factors that may influence the research topic, and that may be very important.

This limited context, which is characterized by quantitative analysis, is a major challenge as well as a limitation.

Another very important thing to know about the possible challenges and limitations of quantitative analysis is the limited depth.

What happens is that quantitative research often focuses on specific aspects of a research topic.

This focus on a specific aspect of the research topic keeps it on the periphery—on the surface.

This type of quantitative research does not have the kind of depth that is desired in a modern context, and it may not be able to provide a very comprehensive understanding of the research topic.

This lack of availability of the results in a very comprehensive manner in quantitative analysis makes it of very limited use.

This is a major challenge.

The third limitation or challenge, whatever you call it, of quantitative research refers to its limited flexibility.

Quantitative research often involves a predetermined research design and methodology, which may limit the researcher's ability to adjust the study design or methods.

As new information and realities emerge, situations are constantly changing.

New knowledge is emerging.

Markets are changing.

They are very dynamic.

What happens is that quantitative research has the potential to become obsolete very fast.

This limited flexibility is a major challenge and limitation of quantitative research.

The fourth challenge of quantitative research refers to the difficulty in measuring complex concepts.

What happens is that quantitative research may not be suitable for measuring complex concepts that cannot be easily quantified.

What are these complex concepts?

Generally, these complex concepts refer to things like capturing the beliefs of the population, or the emotions of the people, or the attitudes of the people, or the feelings of the people.

These may be very crucial to a particular context.

And as I already mentioned, quantitative research is generally very limited in context.

The same difficulty arises in measuring complex concepts.

This is the reason why it has a limited context.

This fourth limitation and challenge of quantitative research makes it not very useful in many contexts, especially in situations where complex concepts are involved.

The fifth challenge of quantitative research refers to the many possibilities and the potential for errors that exist in quantitative research.

What happens is that quantitative research relies heavily on statistical analysis, which can introduce errors in the data analysis as well as the interpretation of the results.

Statistical analysis is based on many assumptions and methods, and they are suitable only for certain situations.

When you apply those statistical tools and analysis, the results may give interpretations that are far from reality.

The potential for errors increases in such cases.

The last very important challenge and limitation—though not the only one, as there are many others—refers to its lack of generalizability.

What happens is that the findings of quantitative research may not be easily generalizable.

Generally, they are not generalizable to other contexts, meaning contexts that are different from the one at hand for the researcher, or even to other populations.

The results of quantitative research are generally suitable only for the context in which the researcher is working and the populations being studied.

This makes the use of quantitative research very limited.

If we also talk about some other issues in quantitative research, these are the problems of biases that are very much inherent.

One of the most important biases generally observed in quantitative research refers to sampling bias.

A sampling bias occurs when the sample used in the study is not representative of the population that is being studied, and that can lead to inaccurate or misleading results.

It is very common in quantitative research.

The second type of bias that is generally seen in quantitative research refers to survey design bias.

What is survey design bias?

It refers to the errors or biases that can be introduced during the survey design phase itself, such as poorly worded questions or limited response options in the design.

Survey designs that are unable to capture the full range of possible responses from respondents also create such biases.

These kinds of biases are very common.

These were some of the challenges and biases that are generally observed in quantitative research.

Regression analysis is a statistical technique used to model the relationship between a dependent variable and one or more independent variables. It helps to understand how changes in the independent variables affect the dependent variable and enables the prediction or estimation of the dependent variable based on the values of the independent variables.

An example of regression analysis can be examining the relationship between a company's advertising expenditure and its sales revenue. Here, the dependent variable is the sales revenue, and the independent variable is the advertising expenditure. By analyzing historical data on advertising spending and corresponding sales revenue, regression analysis can provide insights into the impact of advertising on sales.

To use regression analysis, follow these general steps:

1. Define the Research Question: Clearly define the research question or problem you want to address, along with identifying the dependent and independent variables.

2. Gather Data: Collect relevant data on the dependent and independent variables from a reliable and representative sample or population.

3. Choose the Regression Model: Select the appropriate regression model based on the nature of the data and the research question. Common regression models include linear regression, multiple regression, logistic regression, and polynomial regression.

4. Conduct Regression Analysis: Apply the chosen regression model to the collected data using statistical software or programming languages like R or Python. The analysis estimates the coefficients of the regression equation and assesses the statistical significance of the relationship between the variables.

5. Interpret the Results: Examine the coefficients, p-values, and other statistical measures to understand the relationship between the variables. Interpretation includes determining the direction, strength, and significance of the relationships.

6. Validate and Refine the Model: Evaluate the goodness-of-fit of the regression model and assess its predictive power. Consider diagnostic tests, such as residual analysis and checking for assumptions, to validate the model. If necessary, refine the model by adding or removing variables.

7. Make Predictions or Inferences: Once the model is validated, use it to make predictions or draw inferences about the dependent variable based on the values of the independent variables. These predictions can guide decision-making and help understand the impact of changing independent variables on the dependent variable.

Regression analysis is a versatile tool used in various fields, including economics, finance, marketing, social sciences, and healthcare. It enables researchers and analysts to uncover relationships, make predictions, and gain insights from data. However, it's important to note that regression analysis assumes certain underlying assumptions and requires careful interpretation to avoid misleading conclusions.

In data mining, clustering, association rules, and decision trees are popular techniques used for data analysis and knowledge discovery. Here's a brief explanation of each technique along with examples of its applications:

1. Clustering: Clustering is the process of grouping similar data objects based on their characteristics or proximity. It aims to find inherent patterns and structures within the data. Clustering is an unsupervised learning technique, meaning it doesn't rely on predefined classes or labels.

Example application: Customer Segmentation

In marketing, clustering can be used to segment customers based on their purchasing behavior. By analyzing customer data such as transaction history, demographics, and preferences, clustering algorithms can group customers into distinct segments. This information helps businesses target specific segments with personalized marketing campaigns.

1. Association Rules: Association rules mining focuses on discovering relationships and associations between different items or variables in a dataset. It identifies patterns that frequently occur together and captures dependencies between variables.

Example application: Market Basket Analysis

In retail, association rules can be applied to analyze customer purchasing patterns. For instance, by analyzing sales transaction data, you can identify products that are often bought together. This information can be used for various purposes, such as optimizing product placement in stores, creating product bundles, or generating recommendations for upselling or cross-selling.

1. Decision Trees: Decision trees are hierarchical structures that use a set of rules to make decisions or predictions. They represent a flowchart-like model where each internal node represents a test on an attribute, each branch represents an outcome of the test, and each leaf node represents a class label or a prediction.

Example application: Credit Scoring

Decision trees can be used in credit scoring to assess the creditworthiness of loan applicants. By considering various factors such as income, credit history, employment status, and debt-to-income ratio, a decision tree can be built to predict whether a loan applicant is likely to default or repay the loan. This information helps financial institutions make informed decisions about granting loans.

It's important to note that these are just a few examples, and clustering, association rules, and decision trees have broader applications across various domains in data mining and machine learning.

Linear regression and logistic regression are both widely used techniques in machine learning for quantitative data analysis. Here's an explanation of each concept along with examples of its application:

1. Linear Regression: Linear regression is a supervised learning algorithm used to model the relationship between a dependent variable and one or more independent variables. It assumes a linear relationship between the variables and aims to find the best-fitting line that minimizes the difference between the predicted and actual values.

Example application: Housing Price Prediction

Suppose you have a dataset with information about houses, including features like the number of bedrooms, square footage, and location. By using linear regression, you can build a model that predicts the price of a house based on these features. The model will find the coefficients for each feature, and you can use it to estimate the price of a new house given its characteristics.

1. Logistic Regression: Logistic regression is a supervised learning algorithm used for binary classification problems, where the dependent variable has two possible outcomes (e.g., yes/no, true/false). It models the relationship between the independent variables and the probability of a certain outcome using the logistic function.

Example application: Customer Churn Prediction

Consider a telecommunications company that wants to predict whether a customer will churn (cancel their subscription) or not. Using logistic regression, the company can analyze various customer attributes like call duration, contract type, customer satisfaction, etc., and build a model that predicts the probability of churn. This information can help the company take proactive measures to retain customers.

In both linear regression and logistic regression, it's essential to preprocess the data, perform feature selection, handle outliers, and evaluate the model's performance using appropriate metrics (e.g., mean squared error for linear regression, accuracy, precision, recall, etc., for logistic regression).

It's worth noting that while linear regression is used for continuous outcomes, logistic regression is specifically designed for binary classification problems. However, extensions like multiple linear regression and multinomial logistic regression exist to handle scenarios with multiple outcome classes.

These techniques can provide valuable insights into relationships between variables and make predictions based on historical data, helping businesses make informed decisions in various domains like finance, marketing, healthcare, and more.

Monte Carlo simulations are a computational technique used in quantitative data analysis to model and simulate complex systems or processes with uncertainty. It involves generating random samples from probability distributions and running multiple iterations to obtain statistical estimates or make probabilistic predictions.

The general steps in a Monte Carlo simulation are as follows:

1. Define the problem: Clearly articulate the system or process being studied, the variables involved, and the objectives of the analysis.

2. Model the system: Develop a mathematical or computational model that represents the behavior of the system or process, including the relationships between variables and the sources of uncertainty.

3. Define input distributions: Identify the probability distributions that characterize the uncertainty associated with the input variables in the model. This could include parameters like mean, standard deviation, or correlations.

4. Generate random samples: Randomly draw values from the input distributions to create a set of simulated input scenarios. The number of samples should be sufficiently large to capture the variability and uncertainty adequately.

5. Run simulations: For each set of input values, execute the model or simulation to obtain the corresponding output or outcome of interest.

6. Analyze results: Analyze the outputs obtained from the simulations, such as calculating summary statistics, constructing probability distributions, or conducting sensitivity analyses to understand the behavior and variability of the system.

Example application: Financial Risk Assessment

Monte Carlo simulations are often used in finance for risk assessment and portfolio analysis. For instance, suppose you want to assess the risk associated with a portfolio of investments. You can model the returns of individual assets as random variables with probability distributions based on historical data. By running Monte Carlo simulations, you can generate thousands of possible scenarios of portfolio returns, incorporating the randomness and correlations between assets. This allows you to estimate the distribution of portfolio returns, calculate risk measures like Value-at-Risk (VaR) or Conditional Value-at-Risk (CVaR), and make informed decisions about portfolio allocation and risk management.

Monte Carlo simulations can be applied to a wide range of fields, including engineering, physics, healthcare, economics, and more. They provide a powerful tool for understanding and quantifying uncertainty, assessing risks, optimizing decision-making processes, and exploring the behavior of complex systems under different conditions.

ARIMA (AutoRegressive Integrated Moving Average) modeling is a popular technique in time series data analysis. It is used to model and forecast time series data, which are observations collected sequentially over time. ARIMA models capture the patterns, trends, and dependencies within the data to make predictions about future values.

The acronym ARIMA stands for the three components of the model:

1. AutoRegressive (AR): The AR component represents the dependence of the current value on previous values in the series. It assumes that the current value can be linearly predicted based on a linear combination of past values. The "p" parameter in ARIMA(p, d, q) denotes the order of the AR component, indicating how many lagged values are considered in the model.

2. Integrated (I): The I component accounts for the differencing required to make the time series stationary. Stationarity refers to a stable mean and variance over time. Differencing is performed by subtracting the current observation from the previous one to remove trends or seasonal patterns. The "d" parameter in ARIMA(p, d, q) denotes the order of differencing.

3. Moving Average (MA): The MA component represents the dependency between the current value and the residual errors from past predictions. It assumes that the current value can be predicted based on a linear combination of past residual errors. The "q" parameter in ARIMA(p, d, q) denotes the order of the MA component, indicating the number of lagged residual errors considered in the model.

ARIMA models can be used for various purposes, such as understanding the underlying patterns and dynamics in the data, making short-term forecasts, and identifying anomalies or outliers. The model parameters (p, d, q) are typically determined through model selection techniques like the Akaike Information Criterion (AIC) or the Bayesian Information Criterion (BIC).

ARIMA modeling assumes that the time series is linear and stationary. If the time series exhibits non-linear patterns or non-stationarity, other techniques like SARIMA (Seasonal ARIMA) or more advanced models may be more appropriate.

Overall, ARIMA modeling provides a versatile and widely used framework for analyzing and forecasting time series data across various domains, including finance, economics, sales forecasting, and environmental analysis.

Example:

Suppose you have historical monthly sales data for a particular product over the past few years. Your objective is to forecast the sales for the next few months to aid in inventory management and resource allocation.

Here's how you can apply ARIMA modeling in this scenario:

1. Data Preparation: Organize the monthly sales data in a time series format, where each observation corresponds to a specific time period (e.g., month). Ensure the data is in a suitable format for analysis, including checking for missing values or outliers.

2. Exploratory Analysis: Plot the time series data to visualize the patterns and identify any apparent trends, seasonality, or irregularities. This step helps you gain insights into the data's characteristics and informs the selection of appropriate ARIMA model components.

3. Stationarity: Check if the time series is stationary, which is a key assumption for ARIMA modeling. Use statistical tests or visual inspection to determine if the series exhibits constant mean and variance over time. If it is non-stationary, apply differencing (integration) to make it stationary.

4. Model Identification: Determine the appropriate order of ARIMA components (p, d, q) through model identification techniques. This involves analyzing the autocorrelation and partial autocorrelation plots to identify potential values for p and q. Additionally, the order of differencing (d) can be determined based on the extent of differencing required to achieve stationarity.

5. Model Estimation: Estimate the parameters of the ARIMA model using the historical data. This involves fitting the AR, I, and MA components to the time series. There are various methods for estimating the model parameters, such as maximum likelihood estimation.

6. Model Evaluation: Assess the goodness of fit of the ARIMA model to the data. Use diagnostic checks, such as residual analysis, to ensure that the model adequately captures the patterns and randomness in the data. Evaluate metrics like mean squared error (MSE) or mean absolute percentage error (MAPE) to measure the accuracy of the forecasts.

7. Forecasting: Utilize the estimated ARIMA model to make predictions for future sales. Based on the model, you can generate forecasts for the next few months, along with their associated confidence intervals, providing a range of likely outcomes.

By applying ARIMA modeling in this example, you can gain insights into the underlying sales patterns, identify any seasonality or trends, and generate forecasts for future sales, helping you make informed decisions regarding inventory management and resource allocation.

I'm sure you found this module very useful.

This module focused on quantitative research in the overall market, research that has its own role and importance.

Traditionally, quantitative research has been used extensively in many contexts.

In a broader sense, it is being used almost in conjunction with other methods of market research in a wide, wide range of situations and projects.

It is really important to know in depth about all aspects and the dimensions of quantitative research.

In this module, my idea was to give you a summary and a snapshot of what all are, the popular things, and important things are that you should know in quantitative research. In the next module, we will be talking about qualitative research, which is becoming more and more important nowadays in the modern context of market research.

As the competition is increasing among the companies, they are bound to capture qualitative data that is very, very significant to understanding consumer behavior, market dynamics, and market estimation.

The limitations that are there in quantitative research are compensated for by qualitative research.

This qualitative research is becoming very sophisticated and can use many IT tools, helped by a lot of information that is available on social media 24 seven, 365 days a year.

A lot of information is being posted on the different platforms on the internet, and that provides a lot of input to qualitative research, making it one of the most important areas of market research that will be covered in the next module.

This module focuses on understanding people's attitudes, beliefs, and behaviors through non-numerical data.

Welcome back to module two.

In this course in this module, we are going to talk about qualitative research.

In module one, we discussed quantitative research methods.

In this module, which is the second part of this course, we are going to talk about different methods that use non-numeric data.

The things that are not numerical.

Those are qualitative.

What are the methods to capture that type of data?

That type of phenomenon, that type of experience.

Those we are going to discuss in this particular module.

Therefore, in this module, our focus will be on understanding people's things like attitudes, beliefs, and behaviors using nonnumerical observations.

For example, their behaviors, beliefs, and attitudes are depicted in the comments that they make on social media, or they write something in the articles, books, or texts.

Those are nonnumerical things.

We are going to capture how to capture that information to understand people's attitudes, beliefs, and behaviors.

That is the purpose of this particular module.

So what are the Key concepts and methodologies?

In this module, we will be starting with some key concepts.

These are related to qualitative research and certain methodologies and techniques that are used to capture such kinds of non-numeric data.

The techniques that we will be covering in this module, like focus groups, interviews, or observations, are some of the techniques.

But there are several techniques, including content analysis and document analysis, which we will also be covering in this module.

In this module, our focus will be on understanding the importance of research design, data collection, and analysis.

The importance is greater in qualitative research than in quantitative research because, in quantitative research, numerical data is involved.

The research design, of course, is very important.

Data collection itself is very important.

The analysis becomes very mathematical.

Statistical analysis techniques are used, and your results are delivered by those techniques.

In qualitative research, however, a lot of experience is required in making the research design to get the results, and in the way you collect the data, which is not as straightforward as in quantitative research.

In qualitative research, you have to find different methods of data collection that are representative of what you want to capture about the attitudes, beliefs, and behaviors of people.

And of course, analysis.

There is no one way of analyzing qualitative data.

A lot of experience, judgment, and skills are required to analyze qualitative data.

These are the things that differentiate qualitative research from quantitative research.

Therefore, the importance of research design, data collection, and analysis takes a next level in qualitative research.

When we talk of the nature of methodologies used in qualitative research, the focus of those methodologies is not fixed in nature.

They are based on the choice of the researcher conducting the qualitative research.

Those methodologies generally focus on gathering and interpreting non-numerical data such as words, images, and behaviors.

When we are talking about gathering and interpreting this kind of data, there are no set techniques.

There are no statistical methods.

A lot of judgment and discretion are required on the part of the researcher to arrive at the answers to the questions that are part of the research design, or what the researcher is trying to find out.

The key to such methodologies is to gain a deeper understanding of people's experiences, beliefs, and attitudes.

That is the key.

Talking about the key concepts and methodologies that are so important in qualitative research, we talk about research design.

Research design in qualitative research involves the systematic collection, analysis, and interpretation of data to answer the research questions.

This systematic collection, analysis, and interpretation, as I told you, is quite different from what you do in quantitative research.

The data collection techniques in qualitative research generally use a variety of approaches to collect data, including observation.

Observation is also one of the techniques for collecting qualitative data.

Interviews with people, often using open-ended questions, help to gather the information that you are looking for.

Focus groups, which involve gathering a group of people to work together to generate qualitative insights, are another method.

In those focus groups, activities have to be designed and systematically carried out to arrive at the desired results.

Another technique used in data collection for qualitative research is called document or content analysis.

We will be focusing a lot on document and content analysis in this particular module because many IT tools and resources are available to carry out content analysis, especially for social media posts.

The enormous number of posts that are made 24/7 around the world is are valuable resource for extracting qualitative information about people’s experiences, beliefs, and attitudes.

A lot of IT tools can process the vast amount of information available on social media.

This kind of content analysis through IT tools has become very popular and useful in the present times.

Talking about sampling in qualitative research, sampling in qualitative research uses purposeful sampling.

There is no set method of sampling like in quantitative research.

In qualitative research, sampling involves selecting participants based on specific criteria related to the research design and the questions that are part of the research project.

Sampling in qualitative research is often non-random.

Unlike random sampling in quantitative research, it is based on the researcher's judgment, as I just mentioned.

If we talk of data analysis in qualitative research, it involves a systematic examination of the data, mainly to identify patterns, themes, and relationships.

It may use techniques such as coding, categorization, and triangulation to analyze the data.

Therefore, the techniques used for data analysis in qualitative research are very distinct from those used in quantitative research.

A very important part of the methodologies used for carrying out qualitative research is validation and reliability testing.

Whatever you are doing—the data collected and the analysis techniques used—has to be validated, and reliability has to be estimated.

The aim of this is to ensure that the findings are accurate and reflective of the real situation.

Validity refers to the accuracy and appropriateness of the research methods, data, and interpretations, while reliability refers to the consistency and repeatability of the key findings and insights drawn from qualitative research.

This becomes a very important part because of the inherent nature of qualitative research.

Unlike in quantitative research, the data collected is subjective.

This subjective data, and the results it produces, require a focus on reliability and validity.

The techniques used have to be validated, and reliability tests must be applied to the findings.

We will also discuss different data collection techniques, like

–focus groups,

–interviews, and

–Observation

Not talking about the qualitative research techniques.

That is called the focus group in focus group.

A group interview that brings together a small number of participants, usually between 6 to 10 or a little more.

Also, it is possible to discuss a particular topic or issue.

This kind of activity.

Focus groups.

This technique can provide insights into people's attitudes, beliefs, and experiences.

Definitely, as well as it can generate new ideas and perspectives coming from and via different people with different experiences, knowledge, and backgrounds.

These focus groups can be used to test, for example, new products or services, understand customer needs, or explore social issues.

These are the things that can be done through these focus groups.

If we talk about another technique, that is interviews.

The interview is a one-on-one conversation between the researcher and the participant.

There is a difference in this.

Here researcher is having a one-on-one conversation with the participant, or maybe a few participants.

It can be structured, it can be semi-structured, or it can be unstructured.

Also, with open-ended questions, it can provide in-depth insights into people's experiences, attitudes, and behaviors.

And it can be used to explore sensitive or complex issues.

Generally, sensitive and complex issues are to be captured through interviews rather than focus groups.

Talking about another technique of qualitative research that we talked about is observation, which involves watching and recording people's behavior in a natural or controlled setting.

This technique can provide insights into people's behavior, routines, and social interactions, and it can be used to study social phenomena such as cultural practices or workplace dynamics, or it can also be used to evaluate the effectiveness of interventions or programs.

These are the situations where observation techniques can be very useful.

Another technique that is very widely used in qualitative research is called case studies.

Now, case studies are in-depth analyses of a particular person, group, organization, or a particular phenomenon.

It can provide rich, detailed insights into complex issues or phenomena, and it can be used to explore a range of topics such as organization, culture, social change, or health outcomes.

These are some of the examples.

The applications can be much wider for case studies.

Now, talking about the most popular technique that is being used in qualitative research today, which is content analysis.

Content analysis usually involves analyzing and interpreting textual or visual data, such as news articles, social media posts, or advertising materials.

Content analysis can provide insights into people's beliefs, values, and attitudes, as well as the broader social and cultural context in which they are embedded.

In this particular module, I'm going to take up this example of one case study where we will be doing content analysis using NVivo 12 software.

Here, in this example, we will be studying mental health.

Some research questions are related to mental health.

Those will be discussed.

We will investigate how people use Twitter to communicate about mental health issues.

The research, design, and sampling in this particular study will involve collating a large sample of tweets related to mental health using keywords such as anxiety, depression, or even suicide, and some more keywords can be used over a specific period.

It may be a fortnight, 15 days, or it can be one month for our purpose.

Because of the paucity of time, we will confine ourselves to 15 days of data, and we will capture that data from Twitter.

It can be captured from any other platform also.

Maybe LinkedIn, maybe Facebook, maybe Instagram.

However, we will be capturing the data from Twitter in this particular example.

The methodology that we are going to use will be a systematic examination of the Twitter post.

The data and the aim would be to identify patterns, themes, and relationships related to our study.

Our research questions and content analysis of Twitter posts related to mental health issues will be the main highlights of this particular methodology technique that we will be using.

As I just mentioned, we will be doing coding using the NVivo 12 software.

You can use many other software programs that are available in the market.

Therefore, in this particular study that is related to mental health, we are going to get certain information from Twitter posts, critical information that will be useful for helping out the victims of depression, those who are critical victims of depression, and they can move or graduate into the situation of taking extreme steps, including suicide.

How to prevent taking those extreme steps by the victims of depression?

What kind of information can we gather from the Twitter posts that will be the aim of this qualitative research study?

We will try to find out the experiences, beliefs, and opinions of the different people who are talking about mental health, who are talking about anxiety, who are talking about depression, and the extreme case of suicide.

Those things we will be studying, we will be trying to gather the information that is useful for counseling and helping out the victims of anxiety and depression.

That is the idea of this research design.

And these are our questions in this particular mental health study.

Let's go to this particular mental health study using NVivo 12 software that will illustrate how to do a content analysis of the Twitter posts related to our research designs.

Our questions.

This example deals with the mental health study.

The research topic that we will be using as an example in this module is to investigate how people use Twitter to communicate about mental health issues.

This is the research topic we will take up.

I will demonstrate to you how to conduct this research.

The research design and sampling in this case involve collecting a large sample of tweets related to mental health using keywords like anxiety, depression, and suicide—these kinds of keywords—over a specific period, maybe one month or a fortnight.

We will not keep it a very large volume because that will take a lot of time.

We will be using limited data for this purpose.

The methodology that we will be using will involve a systematic examination of the data to identify patterns, themes, and relationships.

Basically, what we are doing is a content analysis of the Twitter posts related to mental health issues.

The technique that I am going to use in this example is coding, using the NVivo 12 software.

You can use any software.

I am comfortable with this software, so I will be using it.

I will show you how to conduct this research study on qualitative data.

The Twitter posts that we are using here are qualitative.

We will be doing this study, this research, on the sample that is the Twitter posts.

Before we start on this example, let us talk about the detailed methodology.

What is the detailed methodology?

In this methodology, we will be collecting a large sample of tweets related to mental health using the keywords I just mentioned—anxiety, depression, and suicide—over a specific time, and then the tweets can be imported into NVivo 12 software for analysis.

We will be doing that using something called NCapture, which is part of the NVivo 12 software.

You have to add the NCapture extension in Chrome.

Using that, you can capture the data.

Then, researchers can use the NVivo tool to code tweets according to themes.

We will try to find themes like stigma, support, coping strategies, and treatment options.

We will be using NVivo 12 to code the tweets.

After coding, we will see what the different tweets indicate.

We will try to find some inferences in that and, possibly, in whatever data we can collect.

We will be using the software to identify patterns and trends in the data.

If we can find something of that sort—for example, what are the most commonly used hashtags—we can make a list of those hashtags to continuously track the different tweets that are being posted on Twitter using these hashtags.

This will become very useful because hashtags are very important for researchers working on qualitative data from social media posts in today’s world.

We will try to identify those hashtags.

We will also try to identify the most frequently discussed topics.

What are the frequently discussed topics?

What are the themes?

We will try to identify them, and within the themes, if we can identify certain frequently discussed topics, we can make a list of those as well.

That is what we will be doing in this particular research study.

The expected outcomes of this qualitative research study are:

We will possibly try to get insights into how people use Twitter to talk about mental health.

That is very important.

Possibly, this research study could be used to inform efforts to promote mental health awareness and reduce stigma.

We will try to do this, though we have very limited time and data.

How much we can get out of this research, we will see.

We will keep it as an objective—whether we can get some information regarding efforts to promote mental health awareness and reduce stigma.

I am also not very sure, because I will be doing this research study for the first time with you.

In this demonstration, what the outcomes are, I do not know.

But we will try.

That is the objective.

This study could also demonstrate the utility of NVivo 12 software or any other similar software.

Such software will be of a similar type, with a similar methodology.

They will be doing data mining of qualitative data in a very similar way, using coding and related techniques for analyzing large volumes of social media data.

We will demonstrate the use of NVivo 12 software.

That is what we are trying to do here.

Now, let us look at the Twitter data.

Let's try to find out the different Twitter posts related to our research study, related to our focus areas.

Our questions are research questions.

Let's first explore, for example, anxiety.

We use the keyword anxiety and see what the different posts are as far as the keyword anxiety is concerned.

You can see here that very relevant posts are there on the keyword anxiety.

There would be many.

If we go over a very long period, maybe two months or three months, and look at this keyword anxiety, we will get many thousands of Twitter posts.

What we will be doing is going to the advanced search.

Let's go to the advanced search and use this keyword.

When we use the keyword anxiety in an advanced search, we can also use some more words that are very similar.

But anxiety itself is a very clear word.

At present, we do not know the hashtags that are being used for anxiety.

We will leave it here and go straight to the period of the posts.

Let us take April 2023, starting from 10th April, for example, for the fortnight we are trying to capture the data, and let us say 25th April 2023.

Let us search the data.

These are the Twitter posts related to the keyword anxiety in this fortnight.

What we could do: we have already added the NCapture extension in Chrome that can be accessed from here.

We press NCapture.

As you can see here, we have the option of capturing the tweet data as a dataset or web pages as a PDF.

You can do either, but I feel the dataset would be better because you will get a lot of categories. You will know which are the tweets and which are the retweets.

Those things you can find out.

You can have a better classification and tabulation of the data.

Let's keep it as tweets as a dataset.

We write some description—maybe an anxiety-related dataset—and press capture.

When we are doing that, we again go to NCapture and press Show the progress of the capture.

We can see here the progress: 671 to 770 tweets have been captured during the fortnight for this demonstration. We cannot afford to have very large data.

Ideally, you should have at least 10,000 tweets for analysis, but for this demonstration, we will keep it at around 3000 tweets.

We will not go beyond that.

At whatever point enough tweets are captured, we will press stop. 3000 tweets have been captured.

We will stop here.

This data has been downloaded—anxiety-related.

Now let's go back to Twitter and explore another keyword, maybe depression.

Do we have tweets on depression?

We have many tweets there.

What we will do is go to the advanced search again.

We write depression. We go to the period of the tweets.

Again, we take April, from 10th April 2023 to 25th April.

These are the latest dates, actually.

We will search.

Then we will again do the same process of capture.

We write here the depression-related posts dataset two. We capture.

Again, we go to NCapture and show the capture process.

We will be doing the same thing here. We will stop the capture process at around 3000 posts. We will not have very large data because large data takes time.

In any software, when you analyze very large volumes of data, you need time. In this demo, we do not have that much time.

We will keep our datasets to no more than 3000 tweets.

In this case, also, for the keyword depression, we will stop the capture process at around 3000 for demo purposes.

You will have an idea of how this qualitative analysis is being done in this particular research study.

We have stopped here again, and the file has been downloaded.

Now let's go back to Twitter and explore the third keyword, which is suicides.

What we are trying to do is learn about the tweets that are being done on these keywords, these mentions in those posts.

We are trying to capture those tweets, what people are talking about. We will use this qualitative data—the tweet posts. We will try to analyze the qualitative data through NVivo software.

Using the NCapture process, we are capturing.

We will use these three files, three keywords in the software to find out what the different topics are being discussed.

What are the hashtags commonly used in these areas?

We can capture some tweets and analyze them related to stigma, mental health issues, prevention, or treatment—those kinds of things.

Against those things, we will analyze the data.

Let's look at the Twitter posts related to suicides as the keyword.

We go to the advanced search. We write suicides. We specify the period of capture. We press search.

Again, we have got many posts.

What are people posting about?

These are the different posts related to our keyword.

We capture this qualitative information through NCapture and write here the suicide-related posts dataset three. We press capture again. We go to NCapture and show the progress.

Again, we will stop the capture process at around 3000 posts.

That is enough for our demonstration purposes.

Two files on two keywords we have already downloaded from Twitter—qualitative data.

This is our sample.

We are capturing and collecting the data. We will be using this data in NVivo software to do coding. We will code this data according to our themes.

First, we will identify themes.

We will identify topics. We will identify hashtags in these posts.

On this qualitative data, we will try to find out what hashtags are being used.

We will do all that.

Now we have already captured the data on these three keywords.

Let's now use this data in NVivo software.

We will open a new software here.

Our idea is that using the Nvivo 12 software, we will be doing the coding of the data from Twitter posts, qualitative data that possibly contains information that we require—information for finding the ways of preventing depression or suicides, or any treatment options, or if we can reduce certain kinds of stigma if it is there.

We will try to find useful content in these posts using the coding process.

Coding is what coding does.

Coding is done in the software on the qualitative data using the themes, and we already know the themes.

We already know the keywords; we already know in our research design what kind of themes we are talking about.

What we will be doing in this software is first processing the auto code process.

Auto code means the software will try to find certain themes, and it will try to code for all such themes and the topics related to those themes.

From there, we will be only interested in our themes.

We will filter out the topics.

We will filter out the themes that we plan to focus on in this particular research study.

That is our approach.

Here in the NVivo software, we have the option of NVivo Pro and NVivo Plus.

We will be using NVivo Plus because NVivo Plus can do the auto-coding.

Auto-coding will be very useful for us.

It will help us make our process faster.

We will be using NVivo Plus software.

Let's look at Nvivo Plus software.

Here we will import the captured data—that is, the Twitter data.

Twitter posts, approximately a total of 9000 posts, have been captured.

But ideally, you should have at least 30,000 posts—10,000 posts per keyword.

That will be more appropriate for this purpose.

Here is what we will do.

We will start with a blank project, and we will title it as Mental Health Research Project One.

Now we have this software.

Here, we will try to import all the Twitter post data that we have captured.

We will press NCapture and find these three files that we are interested in:

• One related to suicide (the latest file).

• The second is related to depression.

• The third is related to anxiety.

These three files we will import here.

Now all these three files are already imported here.

For example, I will show you that these three files contain approximately 3000 posts each.

Let's press this file related to anxiety.

We double-click this file, and we can see here all the data is there.

Tweet ID is there.

The username is there.

The content of the tweet is also there.

Time. Date. Tweet type. Retweet. Who has retweeted? The number of retweets.

All this information is there.

What we will be doing is creating charts.

In the charts, we will try to find out the common hashtags that are related to anxiety.

These hashtags are our expected outcome because we will try to capture data in the future using these hashtags on social media.

Let us look at the number of references versus hashtags.

As you can see here, we now have the most popular hashtags that are related to anxiety.

The first hashtag is anxiety.

The second is mental health.

The third is depression.

Then some more hashtags are there that may not be very clear, like beat OCD or CBT, and ERP.

Then OCD.

So many hashtags are there.

You can make a complete list of the hashtags.

You can export this chart for your project.

Let's go to our project files, and they will store this information.

We have already saved this chart.

We will now be able to filter out the very common hashtags.

One outcome of this research study is already there.

The same thing for depression also.

For the depression file, we will go to the chart.

Here also we will also change it to a hashtag rather than a username.

Again, we will have another chart that will contain all the common hashtags.

We will export this chart again.

The same thing for the third file that relates to suicides.

We will make a chart out of it.

In this, we will change it to hashtags.

Again, we will export this chart so that we can capture the common hashtags that are being used related to suicides and related to depression.

You can see here all these hashtags are given.

If you want to see, just for knowledge purposes, the location map in the chart, you can change it from a hashtag to a location.

We will get an idea that, for example, if you look at the suicide posts, where are they coming from?

City names are given there.

Country names are given there.

You can see here that India is also there.

New Delhi, India, is also there.

In chronological order, the locations are given from where the posts are coming.

This chart, also, if you wish, you can export.

So many things you can do here.

You can even look at the map.

For example, if you look at the suicide-related posts from India, they are coming from Mumbai, from Ahmedabad, from Sirsa, Haryana, from Lahore, Pakistan, and from New Delhi.

The places are given here in this part of the world.

You can see here anywhere in the world.

You can check from where the maximum posts are coming in this particular dataset for the fortnight.

This is the way you can do some information gathering here.

But our main interest in this software is to identify the themes.

According to our themes, we are required to code the data.

All three files have to be coded according to our requirements.

We will be doing that.

Okay, now we have done some data massaging.

We have tried to find some information.

Now we are with these three files that contain approximately 3000 tweets related to three keywords—anxiety, depression, and suicide—as per our research design.

Our research is about finding qualitative information by coding using themes such as stigma, support, coping strategies, or treatment options.

That is what we are doing.

Let us look at these three datasets and code them according to the auto-code themes.

What this software will do is find out the themes.

It will identify the topics in this data in auto mode and, accordingly, it will code them.

What we will be doing according to this auto-coding is limiting the outcomes according to our research design.

Let the software do this coding in auto mode.

This will be helpful to us in filtering out the things we just discussed.

This coding in auto mode will be on many, many topics.

One advantage of this auto-coding will be that we will come to know the different common topics that people are discussing in the areas of anxiety, depression, and suicide.

That will be useful for us.

We will first look at those themes and what people are talking about, and then we will filter according to our research design and our research questions.

Let's do that.

Let's select all three files.

That becomes a total of 9000 tweets.

As you can see here, there are no codes right now.

There is no coding done on these three files.

The software will be doing it.

We are just asking the software to do auto code.

We'll right-click and press auto code.

In this auto-code wizard, we have the option of identifying themes, and we have the option of identifying sentiments.

In our research study, we do not need to the identification of sentiments.

It is not required.

Why is it not required? Because the topic is such that sentiments may not play that much of a role.

Here we are talking about depression, anxiety, and suicide.

We are trying to find out in this research study what information we can gather to prevent these things.

Sentiments are already negative here.

We need not go into the identification of sentiments, though it can be done automatically.

I will not go into that.

Let us now identify the themes.

I just pressed Next, and the software is doing auto-coding based on the automatic identification of themes in all three sets of data, based on three keywords.

Approximately 9000 tweets are being analyzed for the different themes and topics people are talking about.

That activation is being done in auto mode by the software.

Our aim would be to look at these themes and topics and select those topics that are of interest in this research study.

We will pick up those topics according to different themes and limit our coding to that part.

Using the coding, we will narrow down the enormous amount of qualitative data into smaller data that is of our interest.

That is what we are trying to do in the NVivo software.

Manually, it is very, very difficult.

If you do this manually, it will take months to carry out this kind of activity.

But here, by using Nvivo software, we can do many things.

In this software, we can not only identify themes or sentiments, not also code the data, but also find out frequently used words in this data, which we will try to do.

That will help us further understand the main topics being discussed.

We will do word frequency.

Hashtag analysis we have already done—we have identified several hashtags, the most common ones.

Those will be useful.

But we can also do word frequency in this software.

We can also identify patterns and relationships among certain themes.

Across themes, we can find some associations.

If it is useful for our research study, we will try to do that.

These tweets have simply been captured from Twitter, and we do not know what people are talking about.

Whatever people are talking about, we are trying to capture our information and the answers to our research questions.

That is what we are trying to do.

This software has now analyzed almost all the themes in these Twitter posts—almost 9000 of them.

What we see here is that the themes identified by the software include suicide, https.

HTTPS refers to the various websites where people are discussing these three common keywords in the area of anxiety and depression.

Fries, French fries.

As I understand, fries and French fries refer to comfort food because, in depression, people tend to have certain easy foods, quick foods that are easy to order and get delivered.

French fries, fries, food—those themes are there.

Logically, it looks like these are comfort foods.

Whether they have a good effect on depression and anxiety, I do not know.

Probably, comfort food at least can prevent victims from taking the extreme step of suicide.

Probably it can prevent that.

It also talks about the police.

It also talks about comfort and comfort food.

Committed suicide.

Mental station.

Police station.

Health.

Health is there.

Report.

Friends—what are the roles of friends in fighting depression, reducing anxiety, or preventing suicide?

Surgery.

Autopsy.

Absurd amount.

Fighting.

Sx today.

Reassignment surgery.

Number.

Some of the themes identified and the number of times these themes have been mentioned in these 9000 tweets.

What we will do is try to find out our topics of interest here.

We should also be capturing these themes—what people are talking about and what the topics within these themes are.

By doing so, we will be selecting our interest area as per our research design and what we are looking for in this research.

We will select that.

Why will we select that? By selecting, we will force the software to code the data and filter the data by coding that is of our interest.

For example, when we press suicide, people are talking about committing suicide, suicide rates, suicide blasts, committing suicide, suicide bombing, threatening suicide, suicide bombs, committing suicide today, suicide notes, suicide attempts.

All these different topics people are talking about.

Political suicides.

Suicide Squad.

Male suicide rate.

And so many, many topics people are discussing in suicide.

But our interest is in topics like suicide hotlines.

That is a preventive measure.

It may help in the prevention of suicide.

Here we have topics like suicide prevention policies, youth suicide prevention, suicide survivors, suicide awareness, or survivors of suicide.

These kinds of topics will tell us a little bit about preventive methods or treatment, probably mitigation.

We can quickly make a note of what people are talking about and, at the same time, identify our topics of interest.

Our research focuses on stigma, prevention, and treatment.

Most of the other topics we do not need to look at because they are out of scope for our research.

What we see here is that in this particular list of topics, we have been able to identify certain codes and certain themes of our interest.

This is how we have selected almost all the topics of our interest and are using them.

These topics will go further in this.

Now that we have these three files, they are coded according to our requirements.

That is the requirement of our research: research design and research questions.

These three files are now coded according to that.

They do not contain data that is not of our research interest.

These three files are now very useful for us.

What we can do is simply look at those topics that we have included.

If you remember, we included all these topics, and the coding has been done.

References have been located according to those codes.

What are those codes?

These topics are our codes, and these are all references.

In all, there are not many references.

If you count them, there will be something like 1300 or 1400 references.

Not too many.

Out of almost 9000 tweets, we have about 1000–1200 references.

It is much easier now to find out the answers to our research questions that refer to fighting depression, overcoming anxiety, preventing suicides, reducing stigma, and improving mental health.

If we look at those things, for example, mental health issues, we can see here in this particular issue, mental health issues, we have about 39 references.

They are included in all three files.

If we click here, we will find what these references are.

These references are mentioned here.

From these references, we can jot down many different mental health issues.

Those are going to help us in creating a program for improving mental health.

Using these 39 references, a lot of information can be obtained.

For example, something is mentioned here that social media can contribute to a rise in anxiety, depression, and other mental health issues.

These kinds of statements are there.

From these statements, we can find out the frequencies of these different statements.

If many people are talking about that, then we can include these things for further validity and reliability testing.

At this stage, we can do this qualitative analysis based on these three files that are coded to our requirements.

Similarly, let us look at fighting depression.

We have about 232 references.

If we click here, we will find some references.

Out of these references, some information can be obtained.

For example, somebody says that it is about losing yourself, fighting depression, and then learning how to find yourself again.

It is inspiring.

These kinds of statements can be used in our brochures or in our communication or training programs, where these statements would help in improving mental health and showing a way to the victims of anxiety or depression.

There are many different ways in which we are now in a position to collect inferences and outcomes from the qualitative data.

Coming back to files again, for example, if we look at the depression file here, we can see the coding that we have done—different main codings such as fighting depression, mental health issues, etc.

What we find here is that we have some references.

For example, fighting depression.

We can click here in this chart and find the references there also.

That is another way of finding it.

Similarly, if we click on the suicide file, we can again find many references in this file itself where mental health issues or suicide prevention efforts are mentioned.

All this information is given there, along with how many references are given for each of these topics that are directly related to our research problem.

We can see those topics there.

For example, in this file, it refers to mental health issues.

Here they write many things.

By reading these few reference texts, it is possible to extract information that is of interest to our research problem.

For example, let us have a look at one of the references given here:

“Did you know that 7 in 10 people with autism have a mental health condition? Or that people with autism are at a higher risk of substance misuse and suicide? Learn more about autism and mental health on this website.”

A link is given there.

This is very powerful information, just from some of the tweets from a fortnight.

Suppose we had this data for almost one month or two months.

You can imagine if we had 10,000 or 15,000 tweets per keyword—that means three files, all containing more than 10,000 tweets.

What kind of information could we obtain?

This is a very powerful software that can help you do coding and find out the outcomes that the research requires.

We can do some more things with these files.

Let us now explore the word frequency in these files and see what information we can gather.

Let me remind you that we are also looking at different topics people are talking about in these areas—topics that are of direct concern in the research.

Let us look at word frequency.

We will go to word frequency here.

We are interested in, for example, the top 50 words of a minimum length of six letters.

We can have grouping with stamped words or synonymous words, or stemmed words—anything.

I think in this case, even stemmed words will be all right.

We run a query for the word frequency.

Let's see what results come from all three files.

Here we have the top 50 words.

They include depression, suicide, anxiety, and people’s friends.

You will see that words like “friends” and “people” are coming at the top.

Police are also there at the top.

For example, if you want to understand which words are of our direct focus, we can go to a word cloud.

We can make a word cloud and see which words are most frequent: depressive, suicide, and anxiety.

You can see here “friends” is there.

“Police” is there.

“Someone” is there.

“Committing” is there.

“People” is there.

And the word “miracle” is there.

These words tell you the type of topics people are discussing around our research questions.

This is a very powerful way of finding the topics being talked about.

If these topics are being discussed, we can do deeper digging into these words and the different references mentioned here.

We can extract a lot of useful information that can help in achieving our outcomes—the outcomes expected from this research.

They can be taken care of by these references.

Let me remind you, this particular example was based on smaller data for demo purposes.

If you have a good amount of data, these words and this kind of information will be of much better quality.

In NVivo 12, you can use various commands and features to find relationships and patterns in Twitter post-qualitative data. Here are some key commands and steps you can follow:

1. Importing Twitter data:

   • Start by importing your Twitter data into NVivo 12. This can be done by using the "Import" command and selecting the appropriate file format (e.g., CSV, Excel, or plain text) containing your Twitter data.

2. Coding:

   • Code your Twitter data to categorize and organize it. To start, as discussed in the earlier lectures, you can also use the AUTOCODE command. Reorganize the autocode nodes and topics or create your own nodes (categories) to represent different themes, topics, or concepts that you want to explore in your data. Use the "Create a new Node" command to create nodes.

   • Apply codes to your Twitter data by selecting the desired text and using the "Code" command or dragging and dropping the text onto the relevant node. The same thing can be done to reorganize the auto codes.

3. Querying:

   • Use queries to find relationships and patterns in your Twitter data. NVivo provides several query types, including text search queries, word frequency queries, and coding queries. Here are a few examples:

     • Text search query: Use the "Text Search" command to search for specific keywords or phrases in your Twitter data.

     • Word frequency query: Use the "Word Frequency" command to identify the most frequently used words or phrases in your data.

     • Coding query: Use the "Coding Query" command to find relationships between nodes or coded segments of your Twitter data.

4. Visualization:

   • NVivo offers various visualization options to help identify relationships and patterns in your Twitter data. Some of the visualization features include word clouds (already discussed in earlier lectures), matrix coding queries, and cluster analysis. To access these features, go to the "Explore" tab and select the desired visualization tool.

5. Matrix coding query:

   • Use the matrix coding query feature to identify patterns and relationships between nodes in your Twitter data. This feature allows you to examine the co-occurrence of codes across sources or cases.

6. Framework matrices:

   • Framework matrices are another useful tool for exploring relationships and patterns in your Twitter data. They allow you to compare coded data across different nodes or themes and analyze the intersections between them.

These are some of the key commands and features in NVivo 12 that can help you find relationships and patterns in your Twitter post qualitative data. It's important to note that the specific steps and commands may vary slightly depending on your version of NVivo, so it's always a good idea to consult the software's documentation or help resources for detailed instructions.

Kindly note that the present version of NVIVO is 14, which is almost the same as NVIVO 12. The demo shown is the same for NVIVO 14.

Friends, in this particular module two, we had been talking about qualitative research.

And in qualitative research, we talked about different key concepts, methodologies, as well as techniques, and some of the techniques we tried to use were also used in a very practical manner.

Now, at the end of this module, in the next lecture, you will find one assignment that is based on the example that we had taken in the last lecture, uh, regarding qualitative research.

Similar work has to be done by you in this assignment.

Look at the assignment. Look at the video that is given in the assignment.

And kindly carry out this assignment.

This assignment will be very useful to you.

You will gain a lot of practical knowledge about module two.

Friends.

Welcome back.

Now, at the end of this course, I want to share with you one case study, a very interesting case study that will give you an idea about the significance of qualitative research.

This case study is titled Understanding Consumer Preferences in the Beverage Industry.

This is the title of the case study.

Let me first share with you the background of the company that is referred to in this case study.

This company is a global beverage company.

The hypothetical name of this company is Beverage Company.

It has been experiencing a decline in sales and wants to revitalize its product offerings to meet challenging consumer preferences.

They are not only challenging, they are changing.

Also, because of this change, these consumer preferences are challenging.

The company is known for producing carbonated soft drinks but is increasingly facing competition from healthier and more diverse beverage Options.

This is the challenge the company is facing, which is known for producing carbonated soft drinks, to adapt to the evolving market and regain its competitive edge.

The beverage company decides to conduct a qualitative market research study for three reasons and three objectives.

What are these three objectives?

The first is to gain insights into consumer preferences.

The second is to identify emerging trends.

What are the trends that are emerging in the market in the dynamic market that the beverage industry is in?

And third is to be able to develop a new product strategy.

These are the three objectives of the company to carry out either in-house or through outsourcing qualitative market research.

This is the background of the case study.

Before I go further in this case study, I want to reemphasize the significance of qualitative market research that I have already discussed with you in the context of this case study. I can say that qualitative market research is crucial for understanding the reasons, emotions, and behaviors that drive consumer choices.

These things drive consumer choices, and that is the purpose of qualitative market research to understand these things.

It helps companies like beverage companies gain a deep understanding of their target audience, discover unmet needs, and create products and marketing campaigns that resonate with consumers. This is the help provided by qualitative market research.

By using qualitative research, the company can explore the Y component behind consumer preferences and uncover insights that quantitative data alone cannot provide.

This is the significance that we have discussed earlier, also in this course.

I just wanted to emphasize the significance of qualitative market research in the context of this case study.

Now, let me explain to you how the beverage company went about doing this qualitative market research, and what methodology they used.

First of all, the beverage company hired a market research agency that was highly experienced in qualitative research to conduct this study.

This is the first thing that the company did.

The agency employed a mix of methodologies to collect in-depth data from a diverse group of participants.

Some of the key methods this company used market research company used are.

I'm just explaining these things that they used.

One of the methods that the research company used was the focus group method.

The agency organized focus groups with participants from different demographics, including age groups, lifestyles, and geographic locations, diverse geographic locations, diverse age groups, and diverse lifestyles.

These groups are moderated by skilled researchers.

Those were employed by this market research company, which facilitated discussions in this focus group on various beverages.

Preferences, uses, patterns, and factors that influence their purchasing decisions.

This is the role that was played by these skilled researchers who were part of the market research team.

The second method this company employed included in-depth interviews.

Individual interviews were conducted with consumers who represented specific segments of interest.

These interviews allowed for more personalized and detailed discussions, providing deeper insights into each participant's preferences, experiences, and emotions related to beverages.

What is the role of beverages in their day-to-day life?

And what are the deeper insights into the preferences of the participants and their experiences?

That was the focus of these in-depth interviews.

Another method that this research company used we called ethnographic research.

What is this ethnographic research?

Researchers visited participants' homes or accompanied them during their shopping trips to observe their actual behaviors and decision-making processes while shopping.

This approach helped identify aspects that participants might not easily express during the discussions.

This is the method of observing, giving a more accurate picture of their preferences and habits.

This was done in the ethnographic research by the research company and its team.

Another method that they utilized in this particular context to carry out this qualitative market research study they resorted to online diaries and digital ethnography.

What is this?

Online diaries and digital ethnography.

Participants who were hired in this study were given online diaries or asked to record videos of their beverage consumption on a day-to-day basis, and their beverage consumption experiences.

This approach allowed for real-time insights and helped researchers understand how beverages fit into consumers' daily lives.

Day-to-day life is what is called digital ethnography.

This approach is very popular nowadays for qualitative research.

And that was used.

Another method that we had discussed earlier, also in great detail, was sentiment analysis and social media listening.

We already had a demonstration on this.

The agency Market Research Agency used sentiment analysis tools and social media monitoring to gain additional insights into public discussions, trends, and sentiments surrounding beverages, competitors, and the industry in general.

The beverage industry in general.

This was done in sentiment analysis and social media listening.

Now, based on these methodologies, the market research agency did data analysis and they noted down all the findings.

I am just giving you a summary of this data analysis and findings.

The research agency compiled and analyzed data from all qualitative research methods that we discussed.

Themes and patterns emerged, providing beverage companies with a comprehensive understanding of consumer preferences, needs, and pain points.

What were the pain points?

The analysis highlighted the following key findings.

What were the findings?

The first finding was related to health and wellness.

What is this finding?

Consumers, it was found, are increasingly prioritizing health and wellness, leading to a rising demand for healthier beverage options such as natural fruit juices, functional drinks, and low-sugar alternatives in beverages.

Another finding related to sustainability.

What is this finding an environmentally conscious consumers prefer products with eco-friendly packaging and sustainable sourcing practices by the beverage companies?

This was the second finding of this research.

The third finding related to personalization, customizable beverages, and unique flavors is gaining popularity, especially among younger consumers who seek novel and personalized experiences.

This was the third finding, the fourth finding that emerged from this particular qualitative research study related to convenience on the go and ready-to-drink options becoming crucial for the busy lifestyles of consumers.

This was the fourth finding.

The fifth finding related to brand perception.

What is this finding related to brand perception?

Consumers associate certain beverage brands with specific qualities or lifestyles, and that influences their choices.

This is the finding that was related to a very important area.

That is the brand perception.

So, therefore, what are the implications for beverage companies from this qualitative research study?

And what is the new product strategy that the company can bring out from the findings that we just discussed?

Armed with these valuable insights, the Beverage company developed a new product strategy that aligned with consumer preferences of the type that we discussed in these findings.

They launched a line of natural fruit juices with sustainable packaging that was very, very important, including customizable options through an online platform.

This is the implication, one of the implications, and part of the new product strategy.

Additionally, beverage companies invested in digital marketing campaigns that emphasized health, personalization, and convenience to target different consumer segments.

Effectively, this was another implication and finding of the new product strategy by the beverage company.

What is the conclusion of this case study?

This case study illustrates how qualitative market research plays a pivotal role in helping businesses like that, as beverage companies, understand consumer preferences and adapt to the changing market.

Dynamic market changing preferences of the consumers by employing various qualitative research methods.

The company gains deep insights into consumers' emotions, needs, and behaviors, enabling them to develop innovative and targeted products that resonate with their audience, ultimately driving growth and success in the highly competitive beverage industry.

This was the conclusion of this case study.

This case study I took up to share with you, to illustrate to you all the things that we have learned in this course.

I am sure you found this course very useful.

The examples, the case studies, and the concepts that we discussed in this course.

They were very well researched by me to bring out these contents that are useful to you to help you take off your career in the qualitative and quantitative research field, and build upon this knowledge.

That is the idea in this program on comprehensive training, on market research.

This was the first course that related to qualitative and quantitative research.

There are two more courses in this program that I explained earlier.

And you can review those courses.

The second course deals with market estimation, market dynamics, and market forecasting.

The third course will include topics related to report writing and impactful and very effective report writing.

Of all the activities that you carry out in the comprehensive market research, and also related to the competitive landscaping.

These areas will be covered in the third course of this program.

If you found this course useful, please share the details of this course with your friends, contacts, and colleagues.

And if you have not rated this course, please do so.

Thank you very much.

Now, friends, we have reached the end of this course, and I wish to congratulate you on completing it.

At the concluding session of this course, I would like to emphasize that we, as researchers, are standing at the crossroads of a rapidly evolving business landscape that is changing very fast across the entire world.

The world is changing at an unprecedented pace, driven by technological advancements, shifting consumer behaviors, and emerging market trends. I will briefly touch upon these trends in this session, as they are going to shape the future of market research.

What is happening is that the traditional boundaries between the two methods we discussed in this course—quantitative research and qualitative research—are blurring, giving rise to new approaches that combine the strengths of both techniques. This is a very positive trend that is emerging and will go a long way in shaping the future of market research and the methods we are going to use.

What are the key drivers? If we understand those key drivers, we will understand the changes happening in the market research domain and why they are happening.

One of the key drivers is the unprecedented proliferation of digital technologies, coupled with an explosion of big data. This big data is being created by billions of people around the world, 24/7, 365 days a year, on social media. That is giving market researchers immense power and strength to do things in less time, with more accuracy, and in real time. These are the main key drivers that are going to shape the future of market research.

Now the game is like this for market researchers: they need to harness the power of data and analytics that is already available. Machine learning and, very importantly, artificial intelligence can uncover patterns, correlations, and insights that were previously unimaginable. We had not imagined that we could reach the stage we have reached today. It is really big—the things that can now be done in market research.

This data-driven approach, often referred to as quantitative research with qualitative depth, is going to revolutionize the future of market research. We have to be ready for that.

When we look deeper into emerging markets and new trends in the market research domain, and compare them with traditional research methods, we see that older methods often involved lengthy processes—from survey design and data collection to analysis and reporting. Collecting data in real time is now becoming a reality. Leveraging tools such as mobile surveys, social listening, and predictive analytics, it is possible to collect and analyze data in real time to gain immediate insights.

These methods allow businesses to gain real-time insights into consumer preferences, market trends, and the competitive landscape, empowering them to make data-driven decisions on the fly. This era of real-time insights is already here.

Another very important market trend shaping the future of research is the greater emphasis on understanding the emotional and subconscious drivers of consumer behavior. These drivers go beyond rational decision-making processes that have long been the focus of research. Rational decision-making alone does not provide the complete picture of what shapes consumer behavior and choices.

Techniques such as biometric measurement, neuroscience, and implicit association testing are gaining traction to explore emotional and subconscious drivers of consumer behavior. These empower businesses to develop better strategies, marketing campaigns, products, and offerings. We, as researchers, have to be ready for this new era of emotional and subconscious listening to the consumer’s mind.

Another trend shaping the near future of market research is the increased emphasis on collaboration and co-creation by researchers worldwide. Embracing methods such as crowdsourcing, online communities, and participatory designs is gaining momentum.

Market researchers have long recognized the importance of engaging customers, but traditionally, it was done in a limited way. Now, by embracing methods like crowdsourcing and online communities, businesses can involve customers directly, gain valuable insights, co-create solutions, and build stronger relationships with their target audiences. This collaborative approach not only enhances the quality of research but also fosters a sense of ownership and loyalty among consumers.

In short, I can conclude this course by saying that the traditional core remains. The principles of quantitative and qualitative research are still valid and relevant. But the future of market research will require an adaptive and innovative mindset. By embracing new technologies, leveraging big data, emphasizing real-time insights, understanding emotional drivers, and embracing collaboration, market researchers can stay ahead of the curve and provide invaluable guidance to businesses in navigating the complexities of the modern world.

Thank you for joining this course. Please share the details of this course with your contacts, colleagues, and friends. If you have not yet rated this course, please do so.

Hello to you!

Today I have some appreciative comments for you.

I want to take a moment to congratulate you on fully completing this course.

Your dedication and perseverance throughout this journey have been truly commendable.

Completing a course is no small feat, and I am incredibly proud of the progress you have made and the knowledge you have gained along the way.

I also want to remind you that this course is just one piece of the puzzle.

It is part of our larger VJ Export Mastery Courses series, consisting of 25 courses that I had told you about earlier as well.

These courses are designed to provide you with a comprehensive understanding of the export industry.

On my part, as I had mentioned earlier, I am committed to helping you expand your learning even further by giving you access to more similar courses in the series. On your part, I again have a small request.

Your feedback and rating are incredibly valuable in refining this course and ensuring it remains world-class.

I kindly ask you to leave a rating for the course along with your honest feedback, in case you have not done so yet.

Once again, congratulations on completing the course.

Keep up the fantastic work that you have done in this course, and remember, I am here to support you every step of the way, even after you have completed this course. You can reach out to me anytime for any mentoring or support that you may need.

Thank you very much.

Hello and welcome, and thank you so much for completing this amazing course.

I truly appreciate the time and effort you have invested in developing all types of skills, whether related to export documentation, compliance, international regulations, logistics, or global marketing strategies.

In this short bonus video lecture, I want to share with you a few optional ways you can continue your learning journey, access additional resources, and stay connected with me for future guidance, all while remaining fully compliant with Udemy policies.

If you want to continue receiving educational content on exports, global compliance updates, HS code classification tips, EU/US regulations, logistics strategies, and real-world case studies, you are welcome to connect with me on LinkedIn.

I regularly post export-related insights, free updates, and practical examples that many learners find very useful.

Again, this is completely optional, but if you would like to connect, this is my LinkedIn profile: LinkedIn.com/in/vijeshjain. Along with my activities on LinkedIn, YouTube, Instagram, and many other social media platforms, I frequently share publicly available articles, guidance notes, and updates related to topics such as documentation and compliance, Indian and international customs rules, labeling requirements, global market trends, and policy changes in the EU, USA, UK, and Middle Eastern regions, as well as best practices for exporters.

These free resources can help you stay informed and confident as your export business grows.

For learners who need personalized clarity on specific export matters, such as HS decisions, regulatory compliance, product classifications, labeling reviews, customs queries, international market strategies, or even Amazon US product launch advisory, I also provide such guidance outside Udemy.

If you ever require any of this tailor-made support, you may contact me directly. My email ID is vijesshjain@gmail.com.

Please note that this is only an optional way to reach me outside Udemy, and it is not required to complete this course. It is also not part of the Udemy purchase for this course, which keeps this message fully compliant with Udemy policies.

In addition, I want to cordially invite you to my Discord Knowledge Hub, which has several channels, including the Q&A section, discussion channel, discussion lounge, video lectures channel, and announcement channel. No registration is required to access this knowledge hub or any of these channels.

Simply click the invite link, which is also provided in the resource section of this lecture, and you can access my Discord Knowledge Hub.

Before I close, I want to sincerely thank you once again for joining this course.

I truly hope that this specialized training has added real value to your knowledge base and to your professional journey in international trade.

My mission is to help learners navigate exports more confidently, whether it is compliance, export documentation, import documentation, logistics, or expanding into global markets.

I wish you tremendous success in your future business endeavors, and I look forward to staying connected with you on your path ahead.

Thank you once again, and all the best in your international journey.

Take care of yourself, and see you in another course in this course series.

Dear Learner,

Thank you for completing this course. I appreciate your time, dedication, and interest in strengthening your knowledge of export documentation, compliance, HS classification, logistics, and global market strategy.

This Bonus Section offers optional ways to continue your learning journey, stay connected, and access additional guidance outside Udemy.
Everything here is completely optional, not required to complete the course, and not included in your Udemy purchase, in full compliance with Udemy policies.

1. Connect With Me on LinkedIn (Optional)

If you'd like to follow my educational posts, updates, and insights on global trade, compliance, and international markets, you can connect with me on LinkedIn:

LinkedIn (Optional):
https://www.linkedin.com/in/vijeshjain/

I regularly share free content, industry news, case studies, and compliance tips useful for exporters and global professionals.

2. Visit My Udemy Instructor Profile (Optional)

If you’d like to explore more of my courses on international trade and global business:

Udemy Instructor Profile (Optional):
https://www.udemy.com/user/vijesh-jain-4/

You can browse additional courses, all focused on simplifying global trade and helping professionals succeed in international markets.

3. Optional Personalized Guidance Outside Udemy

If you ever need individual clarity on export documentation, HS code decisions, customs queries, EU/US/UK/UAE compliance, labeling reviews, market-entry strategy, or Amazon USA marketplace compliance, you may reach out to me directly:

Email (Optional):
vijeshjain@gmail.com

Additional Educational Video Resources at YouTube: https://www.youtube.com/@VijeshJain0506

This is only an optional way to connect and is not required for completing the course.

4. Join the Free Discord Knowledge Hub (No Signup Required)

To support continuous learning, I’ve created an open-access Discord Knowledge Hub for all students.
You can join anytime to access discussions, free resources, shared insights, and regular updates.

Join Optional Discord Knowledge Hub (Optional, No Registration Required):
https://discord.gg/wHgqdYe6tz

This community is free, optional, and designed to help learners share knowledge and stay updated with global trade trends.

5. Free Public Resources for Ongoing Learning

I regularly share publicly accessible updates on topics such as:

• HS classification best practices

• Compliance rules for the USA, EU, UK, and UAE

• Labeling and documentation tips

• Customs procedures

• Market-entry insights

• Global trade risks and opportunities

These resources are available on my social channels and are fully free for learners.

Thank You & Best Wishes

Thank you once again for learning with me. I hope this course has added clarity and confidence to your global trade journey. I look forward to staying connected and supporting your continued growth.

Wishing you success in all your international business endeavors.

Warm regards,
Vijesh Jain
Export–Import Consultant & Trainer
VJ Global Academy