Final stats deck Flashcards
Describe the processes involved in Hierarchical regression model building and explain why it is appropriate for this study?
Researchers look through past literature to decide which factors are more consistently influential
Provide examples what past literature says
Due to past literature enabling researchers to rank their importance
Describe the five assumptions of multiple linear regression and, with reference to the appropriate tests provided in the SPSS output, indicate whether the assumptions have been met or violated. (5 marks)?
No multi-collinearity between IVs in model - Largest VIF should be <10, if two >5 then remove one IV
Average VIF should not be considerably >1
- Independence of residuals - Tested with Durbin-Watson statistic
- Statistic can range from 0 to 4, with 2 meaning errors are uncorrelated • Generally values less than 1 or greater than 3 are problematic
- Homoscedasticity of residuals - dots around 0
- Linearity of residuals - dots around 0 get more information from slides in exam
- Normality of residuals - histogram normally distributed
Model answer for a hierarchical report?
Weight and years retired explained 32.1% of the variation in fitness (p = 0.001)
Within this model weight is significant p = 0.002, where as years retired is not 0.681
When adding alcohol the model now explains 59.7% of variation in fitness (p=0.000) and alcohol is highly significant (p = 0.000)
Finally adding the variable of MVPA can explain 64.5% of the variance of fitness (p = 0.030), MVPA was highly significant (p = 0.030)
Therefore the final model predicts the highest percentage of variance of fitness (64.5%) however against previous research years retired isn’t a significant predictor.
Think its actually this:
The sum4dw explains 75.1% of the variance in DXAfat%. The addition of Thigh skinfold to the regression model increases this to 85.8% (ΔR2 = 0.107, p<0.001). These findings imply that the addition of thigh skinfolds to the commonly used sum of 4 skinfolds would improve the accuracy of predicting body fat%.
What is ‘manipulate a combination of both’ analysed by ?
Mixed factorial ANOVA - so when there is within (repeated measures), and between is (independent so difference in the participants)
Independent ANOVA?
Between subjects (different participants) participants don’t do everything
WHat’s repeated measures?
Repeated measures / within subjects (participants do everything)
How to look at table for equation on a mixed factorial ANOVA?
F ( degrees of freedom at top, degrees of freedom at bottom)
P = significant value
np^2 = partial eta value at bottom
Sphericity ?
the assumption that the variances of the differences between conditions are equal?
Mauchly test will indicate a problem if its significant
What’s a post Hoc test?
Following a one way independent ANOVA you can perform pairwise post hoc comparisons to determine which pairs of means are different from one another
• (Normally select this one on SPS) Bonferroni correction
How to find effect size of repeated measures ANOVA?
Go to Tests of Within-Subjects Effects and get the partial eta value from greenhouse geiser
Reporting APA style?
If mauchly test was failed, we will report the greenhouse Geisser corrected degrees of freedom and p values
An example would be for an independent:
F(1.6,11.2) = 3.79, p=.06, np^2 = 0.35, MSE = 13.71
F(greenhouse geiser degrees of freedom, error greenhouse geisser degrees of freedom) = F value, p = 0.06, np^2 = partial eta squared, MSE = mean square of error
Only report pairwise comparisons eg Bonferroni if you have a significant ANOVA
Assumptions of chi-square test?
The expected frequencies should be greater than 5. If an expected frequency is below 5, the result is a loss of statistical power, and may fail to detect a genuine difference
How to report a chi square test when it has 2 variables?
Assumption:
The minimum expected frequency is 14.44 which is greater than 5, so assumption is met
Results (significance of X^2):
X^2 = 25.356, p<0.001, statistically / reject the null hypothesis. Leadership style and performance outcome are associated.
26.3% (10/38) of athletes who had a democratic coach won, whereas 70.4% (114/162) that had a autocratic coach won.
Interpretation/Conclusion
A much greater percentage of athletes won under an autocratic coach than under a democratic coach