Statistics for Health Professionals - A Practical Guide

Population: A population is a complete group of individuals or objects that share a common characteristic, such as all people in a specific country or all plants in a particular region. In statistical analysis, the population represents the larger group that the researcher is interested in studying.

Sample: A sample is a smaller group of individuals or objects selected from a population. The sample is used to represent the larger population and to make inferences about the population based on the characteristics of the sample. Sampling is a crucial aspect of statistical analysis, as it allows researchers to study a portion of the population, rather than the entire population, which can be time-consuming and resource-intensive.

Inferential Statistics: Inferential statistics is a branch of statistics that uses sample data to make inferences about a population. The goal of inferential statistics is to use sample data to make generalizations about the population from which the sample was drawn. Inferential statistics involves a variety of statistical methods and models that allow researchers to estimate population parameters, test hypotheses, and make predictions about the population based on sample data. This type of statistics is often used to test theories, make decisions, and draw conclusions based on sample data, rather than on the entire population.

In healthcare statistics, data can be categorized into several different types based on the measurement scale used to collect the data. The main data types in clinical statistics are:

1. Nominal Data: Nominal data are data that are categorized into non-numeric categories or groups. Examples include gender, race, and blood type.

2. Ordinal Data: Ordinal data are data that are ranked or ordered, but the difference between the categories is not known. Examples include education level and pain severity rating.

3. Interval Data: Interval data are data that have a defined zero point, but the difference between the categories is not known. Examples include temperature and time.

4. Ratio Data: Ratio data are data that have a defined zero point and a defined difference between the categories. Examples include height, weight, and blood pressure.

5. Continuous Data: Continuous data are data that can take on any value within a range. Examples include heart rate, blood glucose level, and body temperature.

Central tendency refers to a single value or summary statistic that represents the "center" of a set of data. The central tendency helps to describe the typical or average value of a data set and is used to represent the entire data set in a single number. The most commonly used measures of central tendency are mean, median, and mode.

The p-value is a statistical measure used in hypothesis testing to determine the significance of an observed result. The p-value represents the probability of obtaining an observed result, assuming that the null hypothesis is true. The null hypothesis is a statement that there is no difference between the observed and expected results.

If the p-value is less than a predetermined significance level (usually 0.05), it is concluded that the observed result is statistically significant and that the null hypothesis can be rejected. This means that the observed result is unlikely to have occurred by chance and suggests a real relationship between the variables being studied.

Probability is a measure of the likelihood of an event occurring. It is expressed as a value between 0 and 1, where 0 indicates that an event is impossible, and 1 indicates that an event is certain to occur.

For example, if the probability of getting heads when flipping a fair coin is 0.5, it means that there is a 50% chance of getting heads and a 50% chance of getting tails.

Probability can be calculated using various methods such as classical probability, empirical probability, and subjective probability. In classical probability, the probability of an event is calculated as the ratio of the number of favorable outcomes to the total number of possible outcomes. In empirical probability, the probability of an event is calculated based on the actual observations made from a sample of the population. In subjective probability, the probability of an event is assigned based on the personal beliefs or opinions of the person.

There are three main types of probability:

1. Classical Probability: It is based on the concept of equally likely outcomes. It is calculated as the ratio of the number of favorable outcomes to the total number of possible outcomes. For example, if you roll a fair die, the probability of rolling a 6 is 1/6 because there is one favorable outcome (rolling a 6) out of six possible outcomes.

2. Empirical Probability: It is calculated based on actual observations or experiments. It is used when the sample data is collected and analyzed to find the likelihood of an event. For example, the probability of a certain drug being effective can be calculated based on the success rate of the drug in clinical trials.

3. Subjective Probability: It is based on personal opinions or beliefs about the likelihood of an event occurring. It is used when the available information is limited, and the outcome is uncertain. For example, a person's subjective probability of winning a lottery can be based on their beliefs about their luck and the odds of winning.

Each type of probability provides different insights into the likelihood of an event occurring, and the choice of which type to use depends on the nature of the problem and the available data.

Probability distribution is a function that describes the likelihood of occurrence of different values in a set of random variables. It maps the possible values of a random variable to the probability of those values occurring. There are several types of probability distributions, including normal distribution, uniform distribution, Poisson distribution, exponential distribution, and others. The shape of a probability distribution depends on the underlying distribution of the random variable and its characteristics, such as the mean, variance, and skewness. Probability distributions play a crucial role in statistical analysis and hypothesis testing, as they allow researchers to make predictions and draw inferences about population parameters based on sample data.

Central Limit Theorem (CLT) is a fundamental theorem in statistics that states that the distribution of the sum (or average) of a large number of independent, identically distributed random variables will tend to be approximately normal, regardless of the shape of the underlying distribution. This means that even if the individual random variables are not normally distributed, the distribution of their sum or average will be close to a normal distribution as the sample size increases. The CLT has important implications for statistical inference and hypothesis testing, as it allows us to use normal distribution-based methods to make inferences about population parameters, even if the underlying distribution is not normal. This is a cornerstone of modern statistical analysis and is widely used in fields such as economics, psychology, and many others. The CLT provides a basis for the development of various statistical models and methods, including confidence intervals, hypothesis testing, and regression analysis.

Normal Distribution, also known as Gaussian Distribution or Bell Curve, is a continuous probability distribution that is symmetrical around its mean and is characterized by its mean, standard deviation, and total area under the curve. It is one of the most widely recognized and widely used probability distributions in statistical analysis. The normal distribution is a mathematical model that describes the behavior of a large number of real-world phenomena, including height, weight, IQ scores, and many others. The normal distribution is a useful model for many phenomena because it can describe the distribution of a wide range of data, from small to large sample sizes, and from tightly clustered to widely dispersed data. It is also a useful model because of its properties, such as its symmetry, its defined mean and standard deviation, and its ability to approximate other distributions through the central limit theorem. These properties make the normal distribution a powerful tool for analyzing data, making inferences about populations, and conducting hypothesis tests.

Poisson distribution is a probability distribution used to model the number of events occurring within a specified interval of time or space. It is a discrete distribution and assumes that events occur randomly and independently. The Poisson distribution is commonly used in fields such as quality control, reliability engineering, and queueing theory, where the number of occurrences of an event is of interest. It is also used in various applications such as predicting the number of calls to a call center, number of visitors to a website, or number of patient visits to a hospital. The Poisson distribution is characterized by a single parameter, the mean (λ), which represents the average number of events per interval.

Chi-square (χ²) is a statistical test used to determine the difference between the expected and observed frequencies in a sample. It is commonly used in hypothesis testing to evaluate the goodness of fit between observed data and a theoretical distribution. The chi-square test can be used to test for independence between two categorical variables or to test for homogeneity between two or more categorical variables.

The test statistic for a chi-square test is calculated by summing the squared differences between the observed and expected frequencies and dividing by the expected frequencies. The calculated test statistic is then compared to a critical value from the chi-square distribution, which is determined by the degrees of freedom and the level of significance desired.

If the calculated test statistic is larger than the critical value, the null hypothesis is rejected, indicating that there is a significant difference between the observed and expected frequencies. On the other hand, if the calculated test statistic is smaller than the critical value, the null hypothesis is not rejected, indicating that the observed and expected frequencies are not significantly different. The chi-square test is widely used in various fields such as psychology, sociology, epidemiology, and education to test hypotheses about categorical data.

Hypothesis testing is a statistical procedure that allows researchers to evaluate the validity of a claim or assumption about a population based on a sample of data. It involves formulating a null and alternative hypothesis, selecting a sample, collecting data, and applying statistical tests to determine if there is evidence to support the alternative hypothesis over the null hypothesis. The results of the hypothesis test are reported in terms of a p-value, which represents the probability of obtaining the sample data if the null hypothesis is true. The p-value is used to make a decision about accepting or rejecting the null hypothesis.

The null hypothesis is a statement about a population parameter that is assumed to be true until proven otherwise. It represents the status quo or the default position and is typically represented by the symbol H0. The null hypothesis is tested against an alternative hypothesis, which represents the opposite of the null hypothesis and is represented by the symbol Ha. The alternative hypothesis represents the researcher's claim or assumption about the population parameter. The goal of hypothesis testing is to determine if there is sufficient evidence to support the alternative hypothesis over the null hypothesis. If the p-value from the hypothesis test is less than a predetermined level of significance, typically 0.05, the null hypothesis is rejected and the alternative hypothesis is accepted.

Type-1 error is the incorrect rejection of a true null hypothesis. It is also known as a false positive. Type-1 error is represented by alpha (α) and is usually set at 0.05, meaning there is a 5% chance of making a Type-1 error.

Type-2 error is the failure to reject a false null hypothesis. It is also known as a false negative. Type-2 error is represented by beta (β) and is inversely related to the power of a test, meaning the higher the power, the lower the chance of making a Type-2 error.

Confidence interval is a range of values, derived from a sample of data, which is used to estimate an unknown population parameter with a certain level of confidence. It represents the range of values within which the true population parameter is likely to fall, based on the sample data, and the degree of precision desired. The wider the confidence interval, the less precise the estimate and vice versa. Confidence intervals are used to make inferences about population parameters, such as the mean, proportion, or standard deviation, based on sample data. The level of confidence associated with a confidence interval indicates the likelihood that the true population parameter falls within the interval, typically expressed as a percentage (e.g. 95% confidence interval).

The Student's t-test is a statistical test that is used to determine if there is a significant difference between the means of two independent groups. It is commonly used in clinical trials, social sciences, and other fields where the sample size is small or the population variance is unknown. The test works by estimating the difference between the means of the two groups and calculating a t-statistic, which is then compared to a critical value from the t-distribution to determine if the difference is significant. The t-test is named after William Sealy Gosset, who used the pseudonym "Student" when publishing his work in 1908.

ANOVA (Analysis of Variance) is a statistical method used to compare the means of two or more groups. It tests the null hypothesis that the means of all groups are equal against the alternative hypothesis that at least one mean is different from the others. ANOVA is used to determine if there is a significant difference between the groups, and to identify which groups are significantly different from each other. It can be used for comparing means of continuous data and is appropriate for one-way and multi-way designs.

Correlation is a statistical relationship between two variables, where one variable tends to change with the other. It measures the strength of the linear relationship between two variables and ranges from -1 to 1. A positive correlation indicates that when one variable increases, the other variable also increases; while a negative correlation indicates that when one variable increases, the other variable decreases. Correlation does not imply causation, as other factors could be responsible for the relationship between the variables. Correlation is useful in identifying trends and patterns in data.

Regression is a statistical method that is used to model the relationship between a dependent variable and one or more independent variables. It helps to analyze the influence of one or more independent variables on the dependent variable. The goal of regression analysis is to find the best-fitting line (or equation) that represents the relationship between the dependent and independent variables. The regression line can be used to make predictions about the value of the dependent variable based on the values of the independent variables. There are several types of regression, including simple linear regression, multiple linear regression, logistic regression, and others.

Hazard ratio (HR) is a statistical measure used in clinical trials and observational studies to compare the occurrence of an event (e.g., death, disease, etc.) between two or more groups over time. It is calculated as the ratio of the hazard (or risk) of an event in one group relative to the hazard in another group. The hazard ratio provides a summary of the effect of a predictor (e.g., a treatment or a risk factor) on the outcome of interest and is commonly used to assess the efficacy of a new treatment or the impact of a risk factor on disease outcome. Hazard ratios greater than 1 indicate an increased risk of the event in the first group compared to the second group, while hazard ratios less than 1 indicate a decreased risk in the first group.

Risk is a measure of the probability that an event will occur in a specified time period. It is the ratio of the number of events that actually occur to the number of opportunities for those events to occur.

Odd ratio is a measure of the association between two binary variables in a population, calculated as the ratio of the odds of the occurrence of the event in one group to the odds of the occurrence of the event in another group. The odds ratio is commonly used in medical research, particularly in case-control studies, to compare the odds of a particular outcome (such as disease) between two groups of subjects.

Life tables, also known as actuarial tables, are mathematical tools used to calculate and present survival statistics in a given population. They are based on the principles of life expectancy and probabilities of death at different ages. Life tables provide a snapshot of the mortality experience of a population, including death rates, survivorship patterns, and life expectancy. They are widely used in demographic research, public health, and insurance to estimate future mortality rates, determine insurance rates, and evaluate the effectiveness of public health programs. Life tables are constructed using data from vital statistics, such as birth and death certificates, and are updated regularly to reflect changing mortality patterns in the population.

Sensitivity and specificity are statistical measures used to evaluate the performance of a diagnostic test or medical screening tool. Sensitivity is the proportion of true positive results (people with the disease who test positive) among all people with the disease. Specificity is the proportion of true negative results (people without the disease who test negative) among all people without the disease. These measures help determine the reliability and accuracy of a test and are useful in determining the potential benefits and drawbacks of a screening or diagnostic tool. Sensitivity and specificity should be considered together, as a test with high sensitivity may result in many false positive results, while a test with high specificity may miss many cases of disease.