Statistics Masterclass: Stats Using SPSS, Excel, R & Python

Explore the scales of measurement and how research questions determine the use of qualitative and quantitative variables, with examples like gender and education level.

Identify the nominal scale as a strictly categorical, mutually exclusive level for discrete or qualitative variables, with examples like doctor or engineer in surveys and analytics.

Explore ordinal scales as ordinal variables with rank ordering, illustrated by education level, tall/short/average height, and socioeconomic status.

Explore how interval scales use equal distances between ranked categories to measure agreement or satisfaction, illustrated by Likert-style rating questions on five-point to ten-point scales.

Examine the debate whether Likert scales are ordinal or interval, including constant spacing assumptions. Conclude that current consensus treats Likert scales as ordinal in social science.

Discover how ratio scales add absolute zero to interval scales, enabling meaningful comparisons and calculations using benchmarks like income, temperature, weight, and exam scores.

Explain four scales of measurement—nominal, ordinal, interval, and ratio—and how they apply to qualitative and quantitative questions, with examples such as gender, education, five point rating scale, price, and income.

Define frequency as the number of times an observation appears and describe frequency distribution as a tabular summary of those counts by class, with gender as an example.

Explore how to build a frequency distribution from a market survey of favorite eating options, determine total respondents, and count each option using manual tally or Excel.

Create frequency distributions in Excel by defining class intervals, choosing appropriate lower limits, using Stur's rule for class count, and generating histograms with Excel's Data Analysis tools.

Learn to create a frequency distribution in SPSS by importing Excel data, recoding into class intervals, assigning codes, generating a frequency table, and plotting a histogram.

Clear the R console quickly with Ctrl+L on Windows or the Mac equivalent, then reuse the shortcut to declutter your workspace, and learn to define a custom clear function.

Familiarize with the R Studio environment, exploring the console, script, environment, history, and plots areas. Learn to write commands, create variables, and manage packages and tutorials.

Enter data in R by typing numbers or importing files, and use the assignment operator—including the equal sign and reverse arrow symbol—to assign values to x and y.

Create numeric and string vectors in R using the assignment operator and c, assign values to score or marks, and build a performers vector with quotes to verify with mean.

Import Excel files into R using both GUI and code, exploring file paths, previews, and the read_excel workflow.

Learn to import an Excel file into R using setwd and read.csv, set the working directory, and import a CSV file to view and analyze data.

Learn how to use the help function in R to get information about functions like table, which creates a frequency distribution, and how to search with two question marks.

Learn to count frequencies in R using the table function, whether data is entered as a list or imported from Excel, and interpret frequency distributions for descriptive statistics.

Explore geometric progression and geometric series, learn how to construct sequences using a scale factor and a common ratio, and prepare to calculate the geometric mean in the next lecture.

Define geometric mean as the nth root of the product of sample values and illustrate with a dataset of ten students' marks, using log transformation and anti log in Excel.

Learn to calculate geometric mean using three methods: nth root, log-based methods, and the geo mean function in Excel, with a practical example and cross-method validation.

Learn to compute the geometric mean in SPSS by copying data from Excel, adjusting decimals, and using the compare means tab options to display the geometric mean and harmonic mean.

Learn to calculate the log of a number manually using a log table. Identify the characteristic and mantissa, and verify the result with Excel.

Use harmonic mean for data that are rates or vary with time. It equals n divided by the sum of reciprocals, including grouped or weighted data.

Explore the properties of the harmonic mean and its use for rates, speed, or ratios. Avoid it with zeros, no change, or large and small disparities.

Learn to compute the harmonic mean for failure-rate data with manual calculation, and validate results using Excel and SPSS demonstrations.

Learn to calculate the mean for grouped data using midpoints and frequencies, applying the formula sum f x over sum f to analyze office absenteeism and guide HR decisions.

Learn to calculate arithmetic mean for grouped data using SPSS and Excel by computing midpoints from class intervals and using compute variable for f x and summing to derive mean.

Learn to calculate the median for discrete data and continuous data with and without class intervals, applying n+1/2 and grouped-data formulas, with practical examples.

Calculate the mode for grouped data by identifying the modal class (50,000 to 60,000) and applying the formula with l1, l2, f-1, f1, and the class width to obtain 54,705.88.

Explore the weighted mean, its formula, and how sample sizes shape pooled averages, with a practical example of three classes and how to compute the combined mean.

Explore positional averages like quartiles, percentiles, and deciles, and see how they divide data into equal parts, with examples of ranking, grading, and interpreting percentiles.

Understand quartiles as cut points that divide a data set into four equal parts, with q1, q2 (median), and q3 defining the lower and upper quartiles and guiding calculation.

Explore how to calculate quartiles for ungrouped and grouped data, focusing on Q1 and Q3 with practical formulas using n, L1, L2, cf, and f.

Learn to calculate quartiles for discrete data using q_k = k(n+1) and for continuous data using q_k = l1 + ((k n/4 − cf)/f) (l2 − l1).

Learn to calculate quartiles for discreet data in an even ungrouped series, using Q1 = (n+1)/4 and Q3 = (3n+1)/4, with Excel sorting and interpolation.

Calculate quartiles for a discrete seven-score data set by ordering the scores, applying the k(n+1)/4 formula to find q1, q2 (median), q3, and q4, and interpret the mid-range insights.

Calculate quartiles for class interval data using Q1 and Q3 with n, cf, f, and L1/L2, illustrated by real estate price data to guide pricing decisions.

Learn how to calculate quartiles in excel using the quartile function on a marks dataset, including q1 and q3, with data ranges and step-by-step guidance.

Learn to calculate quartiles in SPSS through analyze, frequencies, move the variable to the box, select quartiles, and compare with Excel; first quartile 25.25 and third quartile 74.

Explain the merits and demerits of quartiles, showing how they classify data into four parts, provide a positional average, not affected by extreme values, and not suited for inferential analysis.

Understand deciles as a positional average that divides a data set into ten equal parts, learn how to calculate deciles using software, and compare them with quartiles and percentiles.

Explore how deciles are calculated for discrete and continuous data, using d_k = k(n+1)/10 for discrete and d_k = l1 + [k(n+1)/10 - cf]/f times width for continuous class-interval data.

Learn to calculate deciles for ungrouped discrete data using a small 10-observation data set, comparing deciles with quartiles and median, with manual and Excel demonstrations.

Compute the seventh decile for continuous data in a class interval using the decile formula with L1, L2, f, cf, n, and k.

Learn to calculate deciles in SPSS and compare with Excel, using a dataset and cut-point options to create ten equal groups and interpret decile values, including the 80th decile.

Explore percentiles as 100 equal parts of data, compare to quartiles and deciles, and learn their discrete and continuous data formulas for calculating p_k.

Learn how to calculate percentiles for discrete data in ascending order using the provided formula, interpolate between adjacent observations, and verify percentiles such as p85, p90, and p70.

Apply anderson et al.'s percentile calculation method dk = k n / 100, noting decimals use the next value and integers average the kth and k+1th values.

Compare percentile calculations across inclusion and exclusion methods using SPSS and Excel, estimating 70th, 85th, and 90th percentiles, noting slight method-based differences.

Learn to compute percentiles for continuous data using the class-interval formula, including inclusion and exclusion methods, with a real estate example yielding p85 of 56.

Explore measures of dispersion, including standard deviation, to show how data scatter around the mean and how outliers shape the distribution, with stock and mobile price examples.

Identify the two categories of dispersion measures—absolute and relative—and learn how absolute measures retain units while relative measures use unitless coefficients, with practical car price examples.

Explore analytical strategies for dispersion by comparing range, quartile deviation (interquartile range), mean deviation, and standard deviation across individual, discrete, and continuous data sets.

Learn how range, the first measure of dispersion, is the difference between the highest and lowest data points, with weather and temperature examples and a calculation.

Learn to calculate range and coefficient of range for a single series using max and min values, illustrated with temperature example (37 and 26, range 11) and the formula (Imax-Imin)/(Imax+Imin).

Learn to compute the range in discrete series using a dice example, noting that range depends on extreme payouts (I max minus I min) and is unaffected by frequency.

Compute the range for continuous data or the coefficient of range by subtracting the lower limit from the upper limit in class intervals, using real estate price data.

Master range calculation in Excel and SPSS by finding the difference between maximum and minimum values in a data set of 112 observations.

Assess the merits and demerits of the range, noting its simplicity and low data requirements, but its inability to reveal data distribution without supplementary dispersion measures.

Explore quartile deviation, defined as (Q3 minus Q1) divided by two, and the interquartile range. See a practical demo calculating the coefficient of quartile deviation with Excel and SPSS.

Assess the merits and demerits of quartile deviation as a simple measure. Demonstrate how it captures the average gap between q3 and q1, yet is not a true dispersion measure.

Master mean deviation as a dispersion measure and learn to calculate it for range and quartile deviation alongside individual, discrete with frequency, and continuous class interval data.

Compute the mean deviation for ten students' marks using the formula mean deviation equals the sum of absolute differences from the mean divided by n, and interpret the result.

Compute the mean deviation for discrete frequency data by weighting deviations from the mean with their frequencies and dividing by the total frequency; extend to continuous data via class midpoints.

Explore standard deviation as a measure of variability, distinguishing sample and population, using x-bar and mu, with s and sigma, and explaining why n minus 1 (Bessel’s correction) matters.

Define variance as the square of standard deviation, with sample (s^2) and population (sigma^2) forms. Learn the coefficient of variation, the ratio of standard deviation to the mean times 100.

Calculate the sample standard deviation for ungrouped data with s = sqrt(sum(x minus x-bar)^2/(n-1)) and interpret results to guide a four-day work week decision, illustrated by ten employees.

Learn to calculate the sample standard deviation for ungrouped data in Excel by using stdev.s, selecting the data range, and verifying the result matches manual calculations.

Learn to calculate the sample standard deviation for ungrouped data in SPSS by importing data from Excel, using descriptive statistics, and validating results with mean, min, max, and variance.

Calculate population standard deviation for ungrouped data using the n denominator, and compare mean and variability (4.5 and 1.76) with the sample values (4 and 2.16) from a four-day week.

Calculate the population standard deviation in Excel using the STD.P formula on your data range, matching the manual result with a single formula.

Learn how to calculate population standard deviation and compare it with SPSS's sample standard deviation, and note that SPSS provides only sample SD, requiring manual population SD calculation in Excel.

Explore distribution as the arrangement of data over time, illustrated with rainfall and monthly values. See how plotting this distribution reveals patterns for planning and inference.

Explore symmetric and asymmetric distributions by examining how data align around the mean, median, and mode, with examples from sales and IQ scores.

Learn that the normal distribution is a symmetric bell-shaped curve centered at mu, with about 68.3% within one sigma, 95.5% within two sigmas, and 99.7% within three.

Describe skewness as distortion from the normal distribution, and distinguish positive skewness (left-heavy data with a right tail) from negative skewness (right-heavy data with a left tail).

Explore skewness by comparing mean, median, and mode: positive skew means mean > median > mode, while negative skew reverses this; mean–median distance is about one third of median–mode.

Explore absolute and relative measures of skewness as distortion from the mean, noting the mean–mode difference and the three coefficients: Pearson's, Bowley's, and Kelly's.

Compute Pearson's coefficient of skewness from mean, mode, and standard deviation, interpret sign as positive, negative, or zero, and use the stable formula 3(mean minus median)/sigma when mode is unreliable.

Bowley's coefficient of skewness uses quartiles to assess skewness, comparing median distances to Q1 and Q3 with the difference between Q3 and Q1, signaling positive, negative, or zero skewness.

Explore Kelly’s coefficient of skewness, comparing percentile and decile based formulas with Bailey’s and Pearson’s methods, and understand when to apply mean, median, standard deviation, and quartile or percentile data.

download and install the latest version of r 4.0.3 and the free r studio ide, configure language and startup options, and explore the r console and basic commands.

Install PyCharm community edition to start Python development and data analysis, set up a new project, and use the Python console for basic calculations and coding.

Learn to download and install Anaconda, compare it with PyCharm, choose the right installer for your system, set up default Python, and explore the Anaconda Navigator and included tools.

Learn how to access and use Jupyter Notebook for statistics and data science, launch via Anaconda Navigator or command, create a Python 3 notebook, and navigate files, folders, and sessions.

Get familiar with the Jupyter notebook interface by creating and running Python cells, and by copying, pasting, and deleting them, and by writing code.

Run basic Python commands in a Jupyter notebook, using print for strings and simple calculations like 2+3, with outputs displayed in the same cell after pressing control-enter.

Explore logs and exponents, focusing on base and exponent and the equation b^y = x; learn how log transformation smooths curves and links exponential to logarithmic forms.

Explore the log and exponent relationship: log base b of x equals y, with b^y = x, the inverse of exponent. Learn how log transformations smooth skewed distributions in SPSS.

Explore common, natural, and binary logs and how base choice affects results. Use scientific calculators or online tools to compute logs, comparing base ten, base two, and base e values.

Balance information loss and comprehensibility when choosing class interval width and number, aiming for five observations per interval with whole-number widths and five to twenty intervals.

Explore discrete numbers (natural, whole, integers, rational) and real numbers, including irrational numbers like pi and sqrt(2). See how measurement uses these concepts to quantify and analyze data.

Explore continuous, discrete, and open ended class intervals, and distinguish exclusive versus inclusive methods. Apply these boundaries to construct frequency distributions with practical examples.

Activate the Excel data analysis toolpak via file, options, and add-ins, then select the analysis toolpak to access the histogram and other statistical options.