Statistics for Data Analysts and Scientists 2023

Learn how to carry out statistical tests using IBM SPSS (statistical packages for social sciences) with a 30-day trial, focusing on applying tests to business problems.

Access the course resources folder to view all datasets used for solving the business problem, and download the data via the video description for further practice.

Acquire permission from the lead statistics team to use their business samples and use cases in my server, research, and manual, noting explicit usage rights.

Explore inferential statistics by learning how to make inferences from samples about populations, test hypotheses, and understand null and alternative hypotheses, type one error, type two error, and decision rules.

Explore the chi square test of independence, also known as Pearson's chi squared test, to detect relationships between two categorical variables and the necessary assumptions, with a smartphone brand example.

Explore how to perform cross-tabulation with chi-square tests and measure association using Cramer's V in SPSS, interpreting observed and expected counts and percentages.

Interpret the chi-square test results to determine if there is an association between smartphone brand and preferred smartphone, using the Pearson chi-square statistic and p-value to assess significance.

Explore the one sample t-test to determine if a sample comes from a population with a specific mean, and review the assumptions, null hypothesis, and a herring weight example.

Import Erin's bodyweight data, run a one-sample t test against 400 grams, and interpret the results using the test statistic, p value, and 95% confidence interval.

Interpret a one-sample t-test comparing herring weight to 400 g. Report a mean of 369.55 g with p=0.02 and a 95% confidence interval.

Explore the independent sample t-test for comparing means between two unrelated groups on a continuous dependent variable. Review core assumptions of independence, normality, and equal variances with concise examples.

Import and configure data, run an independent samples test to compare mean spending by gender, and interpret 95 percent confidence interval, noting females spend more and the difference is significant.

Interpret results of independent sample t-test by examining the SIG value: a SIG of 0.137 shows no statistical difference in clothing spend between male and female groups at 0.05 level.

Learn how the dependent (paired) sample t-test compares two related measurements on the same individuals, with key assumptions of interval/ratio data, related groups, no outliers, and normally distributed differences.

Analyze reaction times before and after drinking a beer using a dependent samples t-test to compare means, assess significance, and explore correlation with confidence intervals.

Interpret the results of a dependent sample t-test and report a statistically significant increase in reaction time after a beer, from 1165 seconds to 1288 seconds, rejecting the null hypothesis.

Explore Pearson correlation, measuring the strength and direction of association between interval scale variables like age and income, and learn key assumptions: linear relationship, no outliers, and normality.

Execute a Pearson bivariate correlation of age and net monthly income using a thirty-record data set, interpret the correlation matrix, and assess significance with a two-tailed test.

Identify a strong positive, significant correlation between age and net monthly income in euros using Pearson r 0.730 and p < 0.05 from a symmetric matrix.

Learn simple linear regression to predict a dependent variable from an independent variable, clarify correlation not causation, and review assumptions like continuous data, linearity, outliers, independence, and homoscedasticity.

Analyze the relationship between study time and exam scores using simple linear regression, exploring descriptive statistics, correlation, model summary, p-values, and basic plots to validate assumptions.

Explore how study hours predict exam scores using simple linear regression, interpreting R, R squared, ANOVA significance, and the regression equation.

Explore multiple linear regression to predict a dependent variable from two or more predictors while understanding that correlation is not causation, and apply assumptions to real cases like exam performance.

Conduct a multiple linear regression to predict exam score from spent revising and anxiety level, assess model significance with ANOVA, and interpret which predictors are significant in the coefficient table.

Interpret the results of multiple linear regression by examining the R and R-squared values, ANOVA significance, and identifying the significant predictors like hours spent revising and A-level entry point anxiety.