Verifying Statistical Assumptions
Download as PPTX, PDF0 likes484 views
This document discusses statistical assumptions that must be verified when conducting quantitative analysis, including normality, removal of outliers, homogeneity of variance, linearity, and absence of multicollinearity. It provides an overview of how to check each assumption, what the assumptions mean, and what to do if assumptions are not met, such as considering data transformations or non-parametric tests. The document also advertises dissertation consulting services from Statistics Solutions.
Verifying Statistical Assumptions
- 2. Table of Contents Statistics Solutions Services Statistical Assumptions SPSS Demo Summary and Q/A Need help with your dissertation? Call 727-442-4290
- 3. Services Offered by Statistics Solutions • Topic Development • Prospectus or Concept Papers • Introduction Chapter • Literature Review Chapter (identifying articles) • Methodology Chapter (Quantitative/Qualitative). • IRB forms • Data entry templates • Survey Monkey upload • Results Chapter (Quantitative/Qualitative) • Discussion Chapter • Powerpoints for Defense • Journal Publications Need help with your dissertation? Call 727-442-4290
- 4. Normality Assumption Normality Checks Normality is the assumption that the data follows a bell-shaped distribution. Objective tests: Shapiro-Wilk tests and Kolmogorov-Smirnov tests. Statistical significance (p < .05) indicates that the assumption is not met. Visual interpretations: Histogram and normal Q-Q plots If assumption is not supported, may need to shift to non-parametric alternative or consider transformations. Can also try to remove outliers. Need help with your dissertation? Call 727-442-4290
- 5. Removal of outliers Removing outliers Tabachnick & Fidell (2019) indicate that outliers correspond to z-scores exceeding + 3.29 standard deviations from the mean. Can also use boxplots to visually identify outliers. Consider removing these cases to prevent skew in the findings. Need help with your dissertation? Call 727-442-4290
- 6. Homogeneity of Variance Homogeneity of variance Homogeneity of variance verifies that the spread of dependent variables are approximately equal between groups. If assumption is not supported on a t-test, use “equal variances not assumed” test statistic. Need help with your dissertation? Call 727-442-4290
- 7. Linearity Linearity Run a scatterplot between independent and dependent variables. If data follows a linear trend (positive or negative), then the assumption of linearity is supported. Need help with your dissertation? Call 727-442-4290
- 8. Absence of Multicollinearity Absence of multicollinearity This assumption indicates that there is not a strong relationship between the predictors in the study. Use variance inflation factors (VIFs) to test the assumption. If VIFs are below 10 then the assumption is supported. Need help with your dissertation? Call 727-442-4290
- 9. Additional Support Statistics Solutions is a full-service dissertation consulting company providing graduate students timely, editorial support for their dissertations and scholarly projects For information about our services, receive a complementary 30-min consultation available Mon-Fri 9-5 ET Contact us at info@statisticssolutions.com Phone: 727-442-4290