ASSIGNMENT: DNPU 701 Methodology Paper

ASSIGNMENT: DNPU 701 Methodology Paper

ASSIGNMENT: DNPU 701 Methodology Paper

Statistical Analysis
Data analysis is important in research as it translates theoretical information into accurate information. The data analysis part is arguably the most complex, and it offers more insights into a study. Many scholars have developed the habit of undertaking minor courses in data analysis to have the ability to understand the data they collect from the field (Montgomery et al., 2021). Statistical analysis is the process of collecting and interpreting data to understand the patterns and trends shown by the data. The purpose of this write-up is to respond to the statistical question on BMI change, sessions, gender, age, alcohol, and married.

ORDER A PLAGIARISM-FREE PAPER HERE

1. In your own words, describe the variables of interest. In your description, make sure to include at least one measure of central tendency and one measure of variation.
a. The change in the patient BMI is a ratio variable which is mean is 1.64 and a standard deviation is 4.35.
b. Number of sessions is a scale variable where N=30 and the mean is 10.13. the standard deviation is 5.51
c. The age of patients is a scale measurement with N=30, Mean = 53.21 and a standard deviation is 15.39.
d. Race is a nominal scale with the number of samples being 30.
2. Create two APA tables: one for your chi-square test and one for your regression analysis.
Table A: Regression Analysis: Relationship Between bmi-change, sessions, gender and race.
Coefficientsa
Model Unstandardized Coefficients Standardized Coefficients t Sig.
B Std. Error Beta
1 (Constant) -3.954 3.043 -1.299 .210
Number of sessions patient attended .417 .180 .582 2.316 .033
Patient gender is female .971 1.766 .120 .549 .589
race=Asian -.441 2.678 -.040 -.165 .871
race=Black or African American .388 2.265 .043 .171 .866
race=American Indian/Alaska Native -.708 3.082 -.056 -.230 .821
race=White -2.070 2.622 -.203 -.790 .440
Patient age in years .028 .057 .108 .498 .625
a. Dependent Variable: Change in patient BMI (bmi_post-bmi_pre)
b. N =30, CI = Confidence interval *p< .01 ***p< .001

Table B. Chi-Square test
Patient marital status * Patient consumed alcohol in last 12 months crosstabulation
The patient consumed alcohol in last 12 months Total
No Yes
Patient marital status Not Married Count 2 8 10
% within Patient marital status 20.0% 80.0% 100.0%
Married Count 10 7 17
% within Patient marital status 58.8% 41.2% 100.0%
Total Count 12 15 27
% within Patient marital status 44.4% 55.6% 100.0%
X 2=3.844, p=. 26
3. Draw conclusions about RQ1 and 2 based on the analysis you conducted. Address the following for each regression
RQ1: Is there a relationship between the change in patient BMI (bmi_change) and the number of program sessions attended (sessions)?

a. Write the final regression model and the final fitted regression
Regression model
Change in Patient BMI = -3.954 + 0. 417 (sessions) + 0.971 (gender) -.441 (race=Asian) + 0.388 (Race=Black of African Americans) – 0.708 (race = American Indian/Alaska Native) -2.070 (race= white) + (0.028) age
Final fitted regression
Bmi-change= -3.954 + 0. 417 (1) + 0.971 (1) -.441 (1) + 0.388 (1) – 0.708 (1) -2.070 (1) + 0.028 (1)
= – 5.315

b. Provide a substantive interpretation of the slope and main intercept of interest in the first (simple) regression model and the final regression model.
On average, the predicted model shows that without subjecting the patients to any sessions, the change in the patient BMI will be -1.721.
Slope
Simple Regression
Assuming that sessions are treated a unit then the effect on bmi-change would be -1.390. This implies that an increase in the session by a unit does not have a significant change in the patient BMI.
Final Regression model
The output has a p-value that is greater than the level of significance. This shows that theta is no significant relationship between bmi-change and sessions when controlled with gender and age. Therefore, we will undertake the final test by including another variable and observing if there would be any change on the desired relationship.
c. Is there a statistically significant relationship in the final model? Explain using two statistics from your analysis that support your claim
There is no statistical significance between bmi-change sessions, age, and race. The output is explained by p=0.206 (t= -0.453).
d. How does the relationship change from the simple model to the final model as predictors are added?
The relationship does not show significant change. However, the r-squared in the first and the last model differs significantly. This implies that the inclusion of the control variables in the regression model increases the relevance of the model fit.
e. Why would adding predictors change the relationship between sessions and bmi_change?
In the final model, we examined the behavior of the two variables by introducing three control variables. The control variable introduced in this case is gender, age and race. This implies that we will be concentrating on the additional effect that gender, age and race have on the relationship between bmi-change and sessions. The inclusion of gender, age and race as a control variable relies on the assumption that it might play a role in BMI change and, therefore might affect the relationship between the bmi-change and sessions.
f. Make a prediction from the final fitted model
The output has a p-value that is greater than the level of significance. This shows that theta is no significant relationship between bmi-change and sessions when controlled with gender and age. Therefore, we will undertake the final test by including another variable and observing if there would be any change in the desired relationship.
g. Interpret the R-square statistic in the final model
The R-squared value in this final model is 0.38. translating to 38%. This percentage implies that this model is not effective in explaining the relationship between bmi-change and session when treating gender as a control variable. Besides, the SPSS out shows that p= 0.349, which is greater than the 0.05 level of significance. This implies that the relationship is insignificant. Therefore, the outcome is still insignificant even after adding gender as the control variable in the model.
h. How did the R-square change from the first model to the final model as additional predictors were added? What does this tell us about the models?
The R-squared in the first model is 0.176, and the final model is 0.38. This difference explains that the addition of predictors affects the r-squared. The model increase of the predictors’ aid in explaining the relationship between bmi-change and session.
i. Discuss other factors that could influence the outcome
A significant factor that could affect the outcome in the first model is increasing the session units (Shen et al., 2022). Assuming that sessions are treated as a unit, the effect on bmi-change would be -1.390. This implies that an increase in the session by a unit does not have a significant change in the patient BMI. However, if many units treat the sessions, more than 30, the change on the patient BMI would be significant.
j. Is there a clinically significant relationship between sessions attended and the change in BMI?
As explained by the p. =. 21 (t=-1.299), there is no clinical significance relationship between bmi-change and sessions.
RQ2: Is there a difference in alcohol consumption among married patients in the sample?
a. What Chi-square test is appropriate?
The analysis relied on the Pearson Chi-Square test of independence between bmi-change and the dependent variable (sessions).
b. What are the assumptions of the Chi-Square test? How are they met and what should we be concerned about?
This test assumes that both variables are categorical and all the observations are independent.
c. Interpret the results
According to the study by Frankie et al., (2012), Chi-Square test of independence is used to determine whether two categorical variables used in one sample are different from another. The variables used in this sample consist of alcohol consumption and marital status. The output shows that the X2 = 3.844, the p-value is 0.050 at the level of significance of 0.05. Therefore, the sample shows a significant difference in alcohol consumption among married patients.
d. Are the results statistically significant? Are these conclusions robust given the assumptions?
The p= value is 0.50 showing that there is statistical significance. The conclusions are robust given the said assumptions.
e. Are the results clinically significant and is there a reason to be concerned about alcohol consumption among married and unmarried participants in the study?
The difference between alcohol consumption among married and unmarried participants is more than 1%, as explained by the model. There is a reason to be concerned about because the relationship between the variables is significant.

Prepare a brief memo for your CSP chair, Dr. Abby Normal, summarizing the findings of your analysis and proposing a direction for further evaluation. Use statistical jargon sparingly and, when you do, explain what you mean.
To: Dr. Abby Normal
From: Statistician

ORDER A PLAGIARISM-FREE PAPER HERE

Summary of the Findings
The aspect of data analysis explains the concept taken by the researchers moving from a mass of data to applying the data in making meaningful insights from it. Analysis of the collected data aids in the identification of a distinct pattern of a study. Various methods are relevant to more than a single way to approach data depending on the research type. Several methods of approach are applicable depending on the type of study (De Menezes et al., 2021). The sample data have various variables that elicit different meanings. However, our attention was on determining the relationship between bmi-change and sessions. Again, the analysis sought to determine the difference between alcohol consumption among married patients.
The four linear models that have been generated for the first question show that there is not a significant relationship between bmi-change and sessions. The inclusion of various control variables does not also affect the relationship. Again, the r-squared value generated for all the models is low, implying no significant relationship between bmi-change and sessions. The results from these four models are similar, showing that the change in patient BMI is not related to the number of sessions given to these patients. Therefore, attempts to increase the number of sessions to affect the change in the patient’s BMI would not give a significant outcome. In the second question, a chi-square with a simple cross tabulation test was utilized to determine the difference in alcohol consumption among married patients in the sample. The p-value is 0.05, which is less than the level of significance. This outcome shows that there is a significant difference between alcohol consumption among married patients in the sample.
The primary role of this proposal is to outline the outcome of the above statistical analysis by focusing on two main research questions. These questions included finding the relationship between the change in patient BMI (bmi_change) and the number of program sessions attended (sessions), and the second question was finding if there is a difference in alcohol consumption among married patients in the sample. From the above analysis, it is concluded that there is no statistical relationship between the change in patient BMI (bmi_change) and the number of program sessions attended (sessions). However, the analysis also found a statistically significant difference in alcohol consumption among married patients, explained by p=. 050, from the Chi-square test, p= 3.844 (t= -1.07) from the chi-square test.
The sampled data is important in gaining more insights into the relationship between variables. The multiple linear regression tests have turned insignificant. This implies that other non-parametric tests could be carried out to examine the relationship between bmi-change and sessions. Therefore, further investigation is required to determine the linear relationship between bmi-change and sessions fully.
Conclusion
Data analysis plays an important role in allowing the researcher to gain more insights into variables. In this study, the SPSS out put on the multiple and linear regressions have been significant in denying the relationship between change in patient BMI and the number of sessions patients attend. Such kind of information cannot be realized without analyzing the data. The chi-square test has also been significant in determining the difference between alcohol consumption among married patients. Therefore, the analysis has been effective in determining related and unrelated variables.

References
Montgomery, D. C., Peck, E. A., & Vining, G. G. (2021). Introduction to linear regression analysis. John Wiley & Sons. https://doi.org/10.48550/arXiv.2005.10314
Shen, C., Panda, S., & Vogelstein, J. T. (2022). The chi-square test of distance correlation. Journal of Computational and Graphical Statistics, 31(1), 254-262. https://doi.org/10.1080/10618600.2021.1938585
Franke, T. M., Ho, T., & Christie, C. A. (2012). The chi-square test: Often used and more often misinterpreted. American Journal of Evaluation, 33(3), 448-458. https://doi.org/10.1177/1098214011426594