Elements of Inferential Statistics: T-Test - GEO 334 PDF

ELEMENTS OF INFERENTIAL STATISTICS: T-TEST LAB 3: CONDUCTING T-TEST IN SPSS GEO 334: SPATIAL ANALYSIS DR. KHAULA A. ALKAABI OUTLINE  Introduction to Inferential Statistics  What is a T-test?  Assumptions of the T-test  Hypotheses in T-tests  Conducting T-test in SPSS  Common Mistakes in T-t...

ELEMENTS OF INFERENTIAL STATISTICS: T-TEST LAB 3: CONDUCTING T-TEST IN SPSS GEO 334: SPATIAL ANALYSIS DR. KHAULA A. ALKAABI OUTLINE  Introduction to Inferential Statistics  What is a T-test?  Assumptions of the T-test  Hypotheses in T-tests  Conducting T-test in SPSS  Common Mistakes in T-test Analysis INTRODUCTION TO INFERENTIAL STATISTICS  Inferential statistics, which involves making predictions or inferences about a population based on a sample of data.  Hypothesis testing is one of the core methods used in inferential statistics.  Both T-tests and Z-tests are essential tools in hypothesis testing, used to determine if the differences between means (or other parameters) are statistically significant, depending on the sample size and whether the population variance is known. WHAT IS A T-TEST?  T-test as a statistical method used to determine if there is a significant difference between the means of two groups.  Types of T-tests: 1. One-sample T-test: Compares sample mean to a known population mean (e.g., comparing a class average score to a national average). 2. Independent samples T-test: Compares means from two independent distinct groups. (e.g., males vs females). 3. Paired samples T-test: Compares means from the same group at different times or conditions, like a pre-test and post-test scenario. ASSUMPTIONS OF THE T-TEST  The main assumptions for using a T-test: 1. Data must be continuous (interval/ratio scale). 2. The data should be approximately normally distributed. 3. Observations must be independent. 4. For independent samples T-test: Homogeneity of variances between groups. Note: If these assumptions are violated, the results may not be reliable. HYPOTHESES IN T-TESTS Alternative Null Hypothesis Example: Comparing Hypothesis (H1): (H0): No significant test scores between There is a significant difference between two groups of difference between the means. students. the means. COMMON MISTAKES IN T-TEST ANALYSIS  Common mistakes to avoid: 1. Misinterpreting p-values. 2. Failing to check assumptions (e.g., normality, equal variances). 3. Incorrect grouping or variable selection. CONDUCTING ONE-SAMPLE T STATISTICS IN SPSS  Steps for conducting a One-Sample T-Test:  Interpreting SPSS Output 1. Input data into SPSS. 1. In the output, look for the T-value, degrees of freedom (df), and p-value in the One-Sample 2. Go to Analyze > Compare Means > One- Test table. Sample T Test. 2. Determine if the p-value is less than your alpha level 3. In the dialog box that appears, select the variable (commonly set at 0.05) to conclude whether to you want to test and move it to the Test reject the null hypothesis. Variable(s) box. 3. Report Findings:“A one-sample T-test revealed 4. Enter the test value (the population mean you that the sample mean (M = X) was significantly different from the population mean (M = Y), t(df) = want to compare your sample mean against) in T-value, p <.05.” the Test Value box. 5. Click OK to run the T-test. THANK YOU