1. The statistic that measures effect size in Chi square is either the Phi or the Cramer's V. An effect size of .729 that is statistically significant represents a weak relationship between variables.

Group of answer choices

True

False

2. When using the post hoc tests with a one-way ANOVA, you can be assured that an adjustment will be made for multiple comparisons. However, when using the non-parametric equivalent, Kruskal-Wallis for multiple comparisons, individual tests between each pair of two groups are needed to localize the differences in the means. Strictly speaking, the Bonferroni correction could be applied to results to assure that a type 1 error was not made.

Group of answer choices

True

False

3. The statistic that measures effect size in Chi square is either the Phi or the Cramer's V. An effect size of .729 that is statistically significant represents a weak relationship between variables.

Group of answer choices

True

False

4. When using the post hoc tests with a one-way ANOVA, you can be assured that an adjustment will be made for multiple comparisons. However, when using the non-parametric equivalent, Kruskal-Wallis for multiple comparisons, individual tests between each pair of two groups are needed to localize the differences in the means. Strictly speaking, the Bonferroni correction could be applied to results to assure that a type 1 error was not made.

Group of answer choices

True

False

5. With regard to Table 2 on page 691 in the Mayland et al. study on Quality of Care for Dying Patients, please choose the single correct answer from the following choices:

Group of answer choices

The ANOVA F statistic for four composite scales of the ECHO-D questionnaire is presented along with the level of significance

The Hospice vs. Hospital with LCP was a non-significant comparison of mean on communication

The results of the ANOVA in Table 2 requires the researchers to accept the null hypothesis

Ward environment and care were two composite scales that were not statistically significantly different

6. The Mayland et al. study attempts to establish psychometric properties of the ECHO-D instrument (please refer to the procedure on page 689 and Table 1 on p690 in the study). This includes (select all that apply):

Group of answer choices

Test retest reliability of the ECHO-D was also evaluated in the study

The procedure on p.689 also describes efforts to establish predictive validity

Evidence of internal consistency using Cronbach's alpha for each composite subscale

7. Please review Table 2 on p221 of the Darling-Fisher study on RAAPS and choose the best answer from the choices below:

Group of answer choices

No differences were found between users and non-users on percentage of adoloescent patients and years in practice

Race and provider age present major confounders that were not explained in the study

No differences were found between users and non-users on provider and practice setting

There were no differences found when users were compared to non users on race and age

8. Darling-Fisher et al. used a mixed methods approach to developing the RAAPS instrument. Their published paper includes both quantitative and qualitative processes and results.

Group of answer choices

True

False

9. To make sure that a dataset is clean, it is important to take the following step(s). Choose all that apply.

Group of answer choices

Assume that the person who did the original data entry made no errors

Run a boxplot to identify outliers, and then develop a plan (as in Laerd Statistics) to manage them

Visualize the actual entries in the dataset

Run frequencies for all variables to see if there are any values that are not expected

10. Review the following items and decide whether recoding is necessary to result in reverse scoring. Choose all that apply.

These items are from the HIV/AIDS targeted quality of life (HATQOL) instrument mentioned in homework 3. Each item is measured on a 5 point scale with the number "1" indicating "none of the time" and the number "5" indicating "All of the time". Item scores are summed.

Satisfied with physical activity (hat 1)

Physically limited (hat 2)

Control of life (hat 8)

9. Group of answer choices

Recoding makes this easier and more reliable than changing entries in the dataset manually

Satisfied with physical activity and control of life are both phrased positively

An alternative to recoding in order to reverse score, is to use the compute function

Physically limited needs to be recoded (as compared to the other two items presented).

10. A DNP student plans on measuring the impact of a patient education intervention in primary care on pre diabetes and plans to collect HgbA1c values. The student should: (choose the correct answer)

Group of answer choices

Collect A1c data as continuous before and after the intervention

Just collect A1c data as "normal" or "abnormal" categorical data before and after the intervention

11. It is better to collect continuous data for outcome variables, if available, because you can always collapse it to categorical data in SPSS, but you cannot change a categorical variable to continuous without the actual result.

Group of answer choices

True

False

12. A DNP student's project focused on the extent to which a daily surveillance program and launch of an algorithm reduced the incidence of pressure ulcers. Data were collected on 9 components of the algorithm in terms of whether documentation was present in the EHR or not. These included issues like surface areas, nutritional consult, albumin levels etc. Although each component was measured at the nominal level, (0=No; 1=Yes), the student created a summary score on bundle compliance by using "compute" and adding all 9 components together. The resulting variable, called BundleComp resulted in a score between 0 and 9. Data on bundle compliance was compared pre and post to a randomly selected sample of charts. Data on BundleComp were not normally distributed.

What is the correct statistical test?

Group of answer choices

parametric independent sample t test

non parametric 2 independent samples Mann Whitney U

non parametric Friedman test

non parametric k independent sample test

13. A DNP student tested the effectiveness of mindful meditation exercises in nursing staff members on perceived stress using an established tool with reliability and validity, the PSS 10 scale. The compute function in SPSS was used to summarize the PSS resulting in a total score (PSSTotal) that could range from a low of 0 (no stress) and a high score of 40 (high stress). Choose all that apply:

Group of answer choices

PSSTotal is continuous data measured at the nominal level

To test the difference in the means between the pre and the posttest the student needs to use an inferential test

Dependent variable is PSSTotal

Descriptive data can show the mean score of the pre and the post groups, but only an inferential test can determine statistical significance

14. A DNP student tested the impact of a change in workflow in the emergency department on time in minutes to secure an EKG result before and after the workflow change. There were 104 observations in the sample in each the pre and the post groups, but the data were not normally distributed. Preintervention, the mean time in minutes was 43.6 minutes. After the workflow change, the mean time was 13.9. Interpret the following SPSS output.

Independent-Samples Mann-Whitney U Test Summary

Total N 211 Mann-Whitney U 3414.500 Wilcoxon W 9085.500 Test Statistic 3414.500 Standard Error 443.198 Standardized Test Statistic -4.852 Asymptotic Sig.(2-sided test) <.001

Was there a statistically significant improvement? From the choices below, choose all that apply.

Group of answer choices

Yes, the non parametric Mann Whitney U test is the correct inferential test.

No because the student used the wrong test and should have used the independent samples t test

Yes, the Mann Whitney U generates a standardized test statistic, a z score and the p value is well below .05 at <.001

No because the Bonferroni correction was not used