Final Exam — Study Guide

KIN 610: Quantitative Analysis of Research in Kinesiology — Spring 2026

ImportantSPSS Access for the Final Exam

Canvas LMS is currently unavailable. For the final exam on Tuesday, May 12, 2026, you may download and install the free 14-day trial version of IBM SPSS Statistics on your own computer and use it for the exam. Because Canvas is not reachable, the SPSS version inside Canvas will not be available.

Official IBM trial link: IBM SPSS Statistics free 14-day trial

Please download SPSS only from IBM’s official website. Do not use any other download site or any “free SPSS” links from search results, ads, or third-party websites.

ImportantExam Date

Finals Week — May 2026 | RE 276 (Computer Lab) | Full class session (~2 hrs 40 min)

Use this guide to review concepts, practice SPSS procedures, and locate the relevant sections in the Statistics for Movement Science (SMS) textbook before the exam.

How to Use This Guide

This guide maps each exam topic to:

  1. Key concepts you need to know
  2. SPSS procedure you must be able to execute
  3. SMS textbook section — practice navigating to that page now, so you are not searching during the exam

Key topics include Chapter 14 (One-Way ANOVA), Chapter 15 (Repeated Measures ANOVA), Chapter 16 (Factorial ANOVA), and Chapter 19 (Nonparametric Methods). Being comfortable with all four areas will prepare you for any question on the exam.

Practice each SPSS procedure using the course dataset before exam day.

Practice Dataset

Use the dataset below to rehearse every SPSS procedure covered on the exam. It is not the exam dataset — it uses different variables — so working through it will build your skills without revealing exam content.

Download: fitness-training-practice.csv

Variable Descriptions

Variable Description Unit Type
ID Participant identifier — Nominal
Training Intensity group: 1 = Low, 2 = Moderate, 3 = High — Nominal (3 groups)
Lung_Cap Post-test forced vital capacity liters Continuous
Agility_Pre T-test agility time at baseline seconds Continuous
Agility_Mid T-test agility time at week 6 seconds Continuous
Agility_Post T-test agility time at week 12 seconds Continuous

N = 30 (10 per training group). Note: lower agility time = better performance.

Suggested Practice Tasks

Work through these in order — they mirror the structure of the exam:

  1. One-Way ANOVA — Test whether Lung_Cap differs across the three Training groups. Check Levene’s test, report the F-statistic, and make a statistical decision at α = .05.
  2. Post-hoc comparisons — Based on your Levene’s result, run Tukey’s HSD or Games-Howell. Identify which group pairs differ significantly.
  3. Effect size — Report eta-squared (η²) for the Training effect on Lung_Cap. Classify the magnitude of the effect.
  4. APA report — Write a complete APA results paragraph for the one-way ANOVA including post-hoc and effect size.
  5. Mauchly’s sphericity test — Run a repeated measures ANOVA on Agility_Pre, Agility_Mid, Agility_Post. Report Mauchly’s W, χ², and p. Decide whether to apply a correction.
  6. Repeated measures ANOVA — Report the F-statistic for the Time effect using the appropriate row (Sphericity Assumed, GG, or HF). Make a statistical decision.
  7. Bonferroni post-hoc — Identify which pairs of time points differ significantly and describe the direction of change.
  8. APA report — Write a complete APA results paragraph for the repeated measures ANOVA including sphericity decision, F, corrected df (if applicable), post-hoc, and partial η².
  9. Factorial ANOVA — conceptual — A researcher studies the effect of Training intensity (Low, Moderate, High) and Gender (Male, Female) on lung capacity using a 2 × 3 factorial ANOVA. The interaction is significant. What does a significant interaction tell you about the main effects? Why can it be misleading to interpret a main effect when an interaction is present? How would you use simple effects analysis to follow up?
  10. Nonparametric alternative — conceptual — A researcher collects agility data from three small groups (N = 8 each) and finds the distribution is markedly non-normal. Which nonparametric test should they use instead of a one-way ANOVA? What does that test do to the raw scores before comparing groups, and how do you report the result in APA format?
TipCheck Your Work

After completing each task, verify your output makes sense: Lung_Cap means should increase from Low to High training intensity; agility times should decrease (improve) from Pre to Post. For tasks 9 and 10, check your answers against the relevant SMS chapters.

WarningSubmitting SPSS Output on the Exam

Do not paste SPSS tables or charts into your Word answers document. Instead, export your entire SPSS output as an HTML file and submit it separately:

  1. In the Output Viewer, go to File → Export
  2. Select Web Report (*.htm) or HTML (*.htm)
  3. Set Objects to Export to All Visible Objects and click OK
  4. Save as LastName_FirstName_FinalExam_Output.htm and submit via Canvas alongside your Word file

Practice this export step before exam day so it is familiar.

Exam Questions 1 (10 pts)

Key Concepts

  • One-Way ANOVA: compares means across three or more independent groups; extends the t-test logic to multiple groups while controlling the familywise error rate
  • F-ratio: ratio of between-groups variance to within-groups variance; a large F indicates group means vary more than expected by chance
  • Levene’s test: checks whether the variance in the outcome is similar across groups (homogeneity of variance assumption)
    • Levene’s p > .05 → assumption met → use Tukey’s HSD for post-hoc
    • Levene’s p ≤ .05 → assumption violated → use Games-Howell for post-hoc
  • Null and alternative hypotheses:
    • H₀: all group population means are equal (μ₁ = μ₂ = μ₃)
    • H₁: at least one group mean differs from the others
  • Decision rule: if p < α → reject H₀; the group means are not all equal

SPSS Procedure

Analyze → Compare Means → One-Way ANOVA

  1. Move the outcome variable to Dependent List
  2. Move the grouping variable to Factor
  3. Click Options → check Homogeneity of Variance Test (Levene’s) and Descriptive
  4. Click OK

Read the ANOVA summary table: note the F value, df (between, within), and Sig. column.

TipPractice Dataset Variables

Using fitness-training-practice.csv: move Lung_Cap to Dependent List and Training to Factor.

SMS Reference

📖 Chapter 14 — One-Way Analysis of Variance

Practice: Navigate to this chapter and locate the sections on (1) the F-ratio, (2) Levene’s test, and (3) reading the ANOVA summary table in SPSS.

Topic 2 — Post-Hoc Comparisons

Exam Question 2 (10 pts)

Key Concepts

  • Why post-hoc tests are needed: a significant omnibus F only tells you that at least one group differs — it cannot tell you which groups differ
  • Tukey’s HSD: appropriate when Levene’s test is not significant (equal variances assumed); more powerful when group variances are equal
  • Games-Howell: appropriate when Levene’s test is significant (unequal variances); does not assume equal variances or equal group sizes
  • Reading the Multiple Comparisons table: look at the Sig. column for each pairwise row; if p < .05, those two groups differ significantly
  • Direction: note which group has the higher mean difference (I − J column: positive means Group I > Group J)

SPSS Procedure

Analyze → Compare Means → One-Way ANOVA → Post Hoc

  • Check Tukey if Levene’s was not significant
  • Check Games-Howell if Levene’s was significant
  • Click Continue → OK
TipPractice Dataset Variables

Using fitness-training-practice.csv: Lung_Cap is the dependent variable and Training is the factor — same setup as Topic 1. Select the post-hoc test based on your Levene’s result from that run.

SMS Reference

📖 Chapter 14 — One-Way Analysis of Variance

Practice: Find the section on post-hoc comparisons. Confirm you understand why Tukey and Games-Howell are used in different situations.

Topic 3 — Effect Size & APA Reporting

Exam Question 3 (10 pts)

Key Concepts

  • Eta-squared (η²): proportion of total variance in the outcome explained by the group factor; η² = SS_between / SS_total
  • Interpretation — use Cohen’s (1988) benchmarks:
η² Value Interpretation
.01 – .05 Small
.06 – .13 Medium
≥ .14 Large
  • APA format for One-Way ANOVA:
    • Report F, degrees of freedom (between and within), p-value, and η²
    • Example: F(2, 39) = 85.5, p < .001, η² = .81
    • Include group means and SDs in the paragraph or a table
    • Describe post-hoc results: which groups differed and by how much

SPSS Procedure

Eta-squared is calculated from the ANOVA table:

\[\eta^2 = \frac{SS_{between}}{SS_{total}}\]

Some SPSS versions report effect size directly when Effect Size is checked under Options; otherwise, divide SS_between by SS_total from the ANOVA table.

TipPractice Dataset Variables

Using fitness-training-practice.csv: compute η² for the effect of Training on Lung_Cap using the SS values from your Topic 1 ANOVA output.

SMS Reference

📖 Chapter 14 — One-Way Analysis of Variance 📖 Appendix — Effect Size Benchmarks (Eta-squared) 📖 Appendix — APA Reporting

Practice: Find the APA reporting appendix and locate an example of a one-way ANOVA results paragraph.

Topic 4 — Sphericity Check

Exam Question 4 (10 pts)

Key Concepts

  • Repeated Measures ANOVA: tests whether a continuous outcome changes across three or more time points (or conditions) measured on the same participants
  • Sphericity assumption: the variances of the differences between all pairs of time points must be approximately equal; analogous to homogeneity of variance for between-subjects designs
  • Mauchly’s Test of Sphericity: the formal test of this assumption
    • Mauchly’s p > .05 → sphericity assumed → report from the “Sphericity Assumed” row
    • Mauchly’s p ≤ .05 → sphericity violated → apply a correction:
      • ε_GG < .75 → use Greenhouse-Geisser (GG) correction
      • ε_GG ≥ .75 → use Huynh-Feldt (HF) correction
  • ε (epsilon): correction factor; values closer to 1.0 indicate less sphericity violation; GG and HF corrections adjust the degrees of freedom downward, making the test more conservative

SPSS Procedure

Analyze → General Linear Model → Repeated Measures

  1. Enter a Within-Subject Factor Name (e.g., “Time”) and set Number of Levels to 3
  2. Click Add, then Define
  3. Move the three measurement variables (Pre, Mid, Post) to the Within-Subjects Variables boxes in order
  4. Click Options → check Descriptive Statistics and Estimates of effect size
  5. Click OK

Mauchly’s test appears in the Mauchly’s Test of Sphericity table in the output.

TipPractice Dataset Variables

Using fitness-training-practice.csv: set Within-Subject Factor Name to Time, Number of Levels to 3. Move Agility_Pre, Agility_Mid, and Agility_Post (in that order) into the Within-Subjects Variables boxes.

NoteWhat if SPSS shows dots (.) in the output tables?

Dots (.) appear when SPSS cannot compute a statistic. This happens with fabricated or very small datasets where every participant changes by the exact same amount at each time point, making the variance of the difference scores zero. When that error variance is zero:

  • Mauchly’s Test of Sphericity shows “.” — the covariance matrix is singular and the test statistic is undefined.
  • Tests of Within-Subjects Effects (including the Sphericity Assumed row) shows “.” — the F-ratio requires dividing by the within-subjects error (MS_error), and dividing by zero is undefined.
  • Multivariate Tests show “.” — these tests require inverting the error matrix, which is impossible when it is singular.

This will not happen with a real dataset. Real participants always vary in how much they improve, so the error term will always be non-zero and all three tables will show proper values.

If you encounter this in a practice context, it means the dataset is degenerate — there is nothing to report because no valid statistics can be computed from it. Use a dataset with real variability.

SMS Reference

📖 Chapter 15 — Repeated Measures Analysis of Variance

Practice: Navigate to this chapter and locate the sections on (1) the sphericity assumption, (2) reading Mauchly’s test, and (3) the GG vs. HF decision rule.

Topic 5 — Repeated Measures ANOVA: Main Test

Exam Question 5 (10 pts)

Key Concepts

  • Tests of Within-Subjects Effects table: contains the F-ratio for the Time factor; always report from the row that matches your sphericity decision
  • Corrected degrees of freedom: when GG or HF corrections are applied, both the numerator (df₁) and denominator (df₂) df are multiplied by ε, resulting in non-integer values — report these as shown in SPSS
  • Decision: if p < α → reject H₀ → conclude that the outcome changed significantly across time points
  • Why repeated measures is more powerful than between-subjects: individual differences are removed from the error term because the same participants are measured at every time point; this reduces unexplained variance and increases sensitivity

SPSS Procedure

From the output of the Repeated Measures procedure (same as Q4):

Locate the Tests of Within-Subjects Effects table. Read the row for your within-subjects factor (e.g., “Time”):

  • If Mauchly’s was n.s. → use Sphericity Assumed row
  • If Mauchly’s was significant and ε_GG < .75 → use Greenhouse-Geisser row
  • If Mauchly’s was significant and ε_GG ≥ .75 → use Huynh-Feldt row
TipPractice Dataset Variables

Using fitness-training-practice.csv: look for the Time factor row in the Tests of Within-Subjects Effects table (from your Topic 4 output). Select the correct row based on your Mauchly’s result.

SMS Reference

📖 Chapter 15 — Repeated Measures Analysis of Variance

Practice: Find the section on the Tests of Within-Subjects Effects table. Practice identifying the correct row for different sphericity outcomes.

Topic 6 — Post-Hoc Comparisons, Effect Size & APA Reporting

Exam Question 6 (10 pts)

Key Concepts

  • Bonferroni pairwise comparisons: the standard post-hoc procedure for repeated measures ANOVA in SPSS; adjusts p-values for multiple comparisons; identifies which specific pairs of time points differ significantly
  • Partial eta-squared (η²ₚ): effect size for repeated measures; proportion of variance explained by the time factor after removing between-subjects variance; reported in the Tests of Within-Subjects Effects table when effect size estimates are requested
η²ₚ Value Interpretation
.01 – .05 Small
.06 – .13 Medium
≥ .14 Large
  • APA format for Repeated Measures ANOVA:
    • State Mauchly’s test result and sphericity decision
    • Report F with corrected df (if applicable), p-value, and partial η²
    • Example (sphericity assumed): F(2, 82) = 300.1, p < .001, partial η² = .88
    • Example (GG corrected): F(1.87, 76.7) = 48.3, p < .001, partial η² = .54
    • Report Bonferroni post-hoc results: which pairs differed and direction of difference
    • Include means and SDs for each time point

SPSS Procedure

For Bonferroni post-hoc:

In the Repeated Measures dialog:

  1. Click EM Means
  2. Move your within-subjects factor to the Display Means for box
  3. Check Compare main effects
  4. Set Confidence interval adjustment to Bonferroni
  5. Click Continue → OK

For partial eta-squared:

Check Estimates of effect size under Options; partial η² appears in the Tests of Within-Subjects Effects table.

TipPractice Dataset Variables

Using fitness-training-practice.csv: the within-subjects factor is Time (Agility_Pre, Agility_Mid, Agility_Post). In EM Means, move Time to Display Means for, then enable Bonferroni comparisons. Partial η² will appear in your Topic 4/5 output when effect size estimates are checked.

SMS Reference

📖 Chapter 15 — Repeated Measures Analysis of Variance 📖 Appendix — Effect Size Benchmarks (Partial Eta-squared) 📖 Appendix — APA Reporting

Practice: Locate an example of a repeated measures ANOVA APA paragraph in the reporting appendix.

Quick Reference — Key Decision Points

Situation Decision
Levene’s p > .05 Use Tukey’s HSD for post-hoc
Levene’s p ≤ .05 Use Games-Howell for post-hoc
Mauchly’s p > .05 Report Sphericity Assumed row
Mauchly’s p ≤ .05 AND ε_GG < .75 Apply Greenhouse-Geisser correction
Mauchly’s p ≤ .05 AND ε_GG ≥ .75 Apply Huynh-Feldt correction

Quick Reference — SPSS Menu Paths

Analysis SPSS Path
One-Way ANOVA + Levene’s test Analyze → Compare Means → One-Way ANOVA → Options: Homogeneity of Variance Test
Post-hoc (Tukey / Games-Howell) One-Way ANOVA dialog → Post Hoc
Eta-squared One-Way ANOVA → Options → Effect Size (or calculate from SS_between / SS_total)
Repeated Measures ANOVA Analyze → General Linear Model → Repeated Measures
Mauchly’s test Appears automatically in Repeated Measures output
Bonferroni post-hoc (RM) Repeated Measures → EM Means → Compare main effects → Bonferroni
Partial eta-squared Repeated Measures → Options → Estimates of effect size

SMS Textbook — Sections to Review Before the Exam

Practice navigating to each of these pages in the SMS textbook before the exam day so you are not searching during the exam.

Topic SMS Chapter / Appendix
One-Way ANOVA (F-ratio, Levene’s, post-hoc) Ch. 14
Repeated Measures ANOVA (sphericity, corrections) Ch. 15
Factorial ANOVA (main effects, interactions, simple effects) Ch. 16
Nonparametric Methods (Kruskal-Wallis, Friedman’s, chi-square) Ch. 19
Effect size benchmarks (η², partial η²) Appendix
APA reporting examples Appendix
SPSS: One-Way ANOVA Appendix
SPSS: Repeated Measures ANOVA Appendix