Simple ANOVA 2

Question 1

assumptions of a repeated measures ANOVA

Answer 1

• DV should be at the scale level
  
  • Data should be normally distributed
  • Sphericity (similar to equal variances, though a bit more complex)
  • CONDITIONS must be DEPENDENT
    ○ For example, the same person should be in multiple groups
    If someone is missing from a condition, they cannot be used

Question 2

how to deal with sphericity

Answer 2

• Test and correct
  ○ Test for a violation of sphericity (typically using Maulchy’s test for sphericity)
  ○ If sphericity if violated, correct for it (typically using Huynh-Feldt for minor violations, and Greenhouse-Geisser for more major violations
  ○ Protects Type I error rate, hurts power
  
  • Avoid it
    ○ Use a different approach that does not assume sphericity
    Specifically, linear mixed (effects) models, which we’ll use here

Question 3

what to do on R

Answer 3

• Step 1: Install (if necessary) and load in necessary packages
  
  • Step 2: Load the data
  • Step 3: Look at and format the data
  • Step 4: Test assumptions
  • Step 5: Run the “ANOVA” (linear mixed model)
  • Step 6: Get the important information
  • Step 7: Run post-hoc tests (non-directional hypothesis) and/or planned comparisons (directional hypothesis)
    Step 8: Get the important information

Question 4

report the results (non directional)

Answer 4

‘A linear mixed effects model was conducted to compare the effect of [IV: in this case music type] on [DV: in this case anxiety symptoms]. There was a statistically significant effect of music type on anxiety symptoms (F(2,54) = 5.66, p = .006, ηp2 = .17). Bonferroni adjusted post-hoc tests demonstrated a significant difference between Coldplay and Eminem (p = .006), but not Coldplay and Whale (p = .982) or Whale and Eminem (p = .078).

Question 5

report the results (directional)

Answer 5

‘A linear mixed effects model was conducted to compare the effect of [IV: in this case music type] on [DV: in this case anxiety symptoms]. There was a statistically significant effect of music type on anxiety symptoms (F(2,54) = 5.66, p = .006, ηp2 = .17). Planned contrasts demonstrated a significant difference between Coldplay and Whale and Eminem (t(54) = -2.46, p = .017, ηp2 = -1.14). There was also a significant difference between Whale and Eminem (t(54) = -2.29, p = .026, ηp2 = -0.61).