Research on Dissertation Minimum Standards Rubric.

Dissertation Minimum Standards Rubric

• To be completed by the chair, committee member, and University Research Reviewer (URR) at the proposal and final dissertation stages. This will be sent to the URR as part of the required review documents.
• All Indicators are required. If review suggests that all parts of the indicator are not complete, the appropriate score is that the indicator is Not Met.

Grayscale denotes items relevant to the final dissertation only. All reviewers must rate each area as Met in order for the document to be considered as having met minimum standards.

Date: (click here and type today’s date ) 03/29/15

Student’s Name: Christian L. Way Student ID (for office use only) — A00024558
School: (click here and pull down to select school name ) PSYC

Committee Members’ Names:
Chairperson Dr. William Disch
Member Dr. Jay Greiner
University Research Reviewer Dr. Karen Gil

Stage of the Rubric: Before Dissertation Oral Name of member providing this evaluation: Gil
Overall Score: Not Met

Rubric Items Met/Not Met Comment History
1. The problem is clearly articulated, worthy of doctoral level research, and within the scope of the discipline. Met

2. The work is grounded in a focused application of the relevant theories or conceptual frameworks. Met

3. The literature review is exhaustive and reflects mastery of the current state of knowledge in the discipline related to the area of research. Met

4. The problem statement, purpose, research questions, and/or hypotheses, design, and methodology are consistent with the state of knowledge development in the discipline described in the literature review. Met

5. The research design and methodology, including issues of sampling, sample size (quantitative and mixed-methods studies), participant numbers (qualitative and mixed-methods studies), instrumentation, data collection, data analyses, and procedures, are appropriate to answer the research questions and/or test hypotheses. Met

6. The problem statement, purpose, research questions and/or hypotheses, design, and methodology are consistent and aligned. Met

7. The study describes implications for positive social change at the appropriate levels—individual, community, and/or societal (proposal and final study) Met

8. Results are accurately presented and are aligned with the research questions and/or hypotheses, design, and analysis. Not Met

Experiment 1:
Page 69. A power analysis should not be conducted for a multiple regression analysis since this was not used. The power analysis should have been conducted for the independent samples t test used. Additionally, the solution to conducting an underpowered study is not to use a non parametric test. This does not increase the power of the test. This section should be re-written to describe a power analysis conducted for an independent t test. For example, the minimum required sample size necessary for yielding a statistical power of .80, to detect a medium effect size of 0.70 with a Type I error held at the .05 level of statistical significance (two-tailed test), was determined to be 34 (for each group) for an independent samples t test. The sample size was increased slightly (to 40 per group) to account for possible attrition.
Remove Table 1 as it is redundant with Table 2. If this information is relevant, please conduct a Chi Square analysis of data in Table 2.

Table 8: The median number is incorrect.

Tables 15 and 16. There is no rationale given for conducting individual t tests for each of the components (vault, bar, beam, floor) and unlikely to be enough power to conduct these kind of sub group analyses. Please just present and discuss the Total score (as described in the Research Question).

Experiment 2.

Page 87. This is not the correct way to conduct a power analysis: “Each subgroup of the independent variable has a sample size of 10, which is adequate, since it is recommended to have a minimum sample size that is more than the number of dependent variables.” It would acceptable to state that a power analysis of an ANOVA with 3 groups with a large effect size (0.28) required 10 participants/group.

The correct test to perform would have been a two way repeat measures ANOVA for each dependent measure. This would have tested whether there was a significant effect due to Intervention (difference between pre and post test scores), an effect due to Group, and an interaction between Intervention and Group. In Tables 22, 23, 25, 26, 28 and 29 results from 23 t tests are presented. It is not acceptable to conduct an analysis of pre- and post-tests among 30 participants in 3 groups using 23 statistical tests.
It appears that there should be 3 dependent measures (anxiety, self-efficacy and locus of control) not the 7 variables presented (see for example Table 21). The variables should have been combined, for example the two Anxiety scores (cognitive, somatic) into one Anxiety score (CS), and the SE vault (V), bars (B), floor (F) into one SE score for V, B, F). The ANOVAs should then have been conducted using the combined scores (e.g. anxiety, SE(VBF) or Anxiety (CS) as the dependent measure.

In fact, here is the stated purpose from page 85, which identifies 3 dependent measures: “It was hypothesized that the autogenic experimental group will yield a significant reduction in perceived levels of PCA (as measured by Competitive State Anxiety Inventory-2C), increased levels of perceived self-efficacy (as measured by Sport Specific Self Efficacy Scale), and increased levels of perceived internal locus of control (as measured by Nowicki-Strickland Locus of Control Scale), between pre- and post-testing as compared to guided imagery and control group.”

So, 3 two way repeat measures ANOVA rather than 23 t tests should have been conducted. Post hoc t tests should then be conducted.

Not enough information is provided concerning the MANOVA, presented on pages 93 – 95 to understand what was done (or why the test was conducted). This would not seem to be an appropriate test to conduct, although since the reasoning was not provided, it is not possible to make this statement conclusively. If this section is retained it needs to be significantly expanded to explain what was done and why, however, if the 2 way repeat measures ANOVA are conducted then this section will likely not be necessary.
Summary statements and Chapter 5 should be written in terms of the results from the 2 way ANOVA.
9. Conclusions, limitations, and recommendations are clearly described, are appropriate to the study scope, and are integrated into the state of knowledge described in the literature review. Met