## Framingham baseline data | Bio statistics Sample | GL31

You can download the solution to the following question for free. For further assistance in  Statistics assignments please check our offerings in Statistics assignment solutions. Our subject-matter-experts provide online assignment help to Statistics students from across the world and deliver plagiarism free solution with free Turnitin report with every solution.

(ExpertAssignmentHelp do not recommend anyone to use this sample as their own work.)

### Question

Taking the Framingham baseline data files for men and women as random samples of the US population, address the research question:

Does education level predict total cholesterol in the US males and/or US females after correcting for age?

1. ## Provide your appraisal of the strengths and weaknesses of the presentation of the statistical material in Winkleby et al, 1992

Strengths: In my opinion, the primary strength of Winkleby et al’s description of statistical material is that it is completely to the point and at the same time quite comprehensive. The details about the samples are well presented with having important information like the span of age of the population considered.

. Also, what was done with the unemployed samples is well stated. As a non-responded questionnaire is quite commonly encountered in these kinds of studies, the assumption to handle those cases is adequately described.

Thirdly, all the independent explanatory variable are separately described to ensure that the choice of analytical framework gets justified. The coding method for dependent variable is also discussed to substantiate the analysis.

Weaknesses: Along with these strengths, I found quite a few weaknesses of this portion which should have been taken care of.

First of all, the total number of participants in each of the cross-sectional study was not mentioned. Furthermore, the absolute number of dropped unemployed participants in various categories are given, their percentage in the total number of participants, however, is not clearly mentioned. If this percentage was found to be significant, then alternatives to dropping those participants should have been considered. Moreover, the gender distribution of those dropped participants is not at all highlighted.

Secondly, how partially completed questionnaires are dealt is not described. Generally, entries with ‘NA’ can affect the regression substantially, hence the mechanism to handle those entries should have been described sufficiently.

Thirdly, in my opinion, the description of the analytical methods is not adequate. The different types of regression models used in the study are not discussed in detail. The basic equation for those models also should have been discussed in the method section of the paper. Moreover, the rationale behind choosing those methods also be stated clearly in this section. Even, the software used for conducting the statistical analysis is also not mentioned.

Fourthly, ‘homemaker’ category is dropped from the multivariate analysis which actually will add up to 1/3 of women population. The justification for dropping that category is not provided, and any other alternative is also not considered.

Finally, it is not mentioned, whether in the study (both in the data collection process as well as in the analysis) any ethical code and conducts were conformed or not.

1. ## Provide the description and results of your descriptive analyses (100-150 words, 8 marks)

Total 1934 men and 2490 women have participated in the survey. The sample contains data about age, educational qualification, and total cholesterol. Educational qualification is an ordinal variable with 4 different levels: 0-11 years, high school diploma,

some college, and college degree. Total cholesterol and age are continuous variable. Table I summarizes the mean and SD for age and total cholesterol, and total frequency and percentage of total participants for educational qualification. According to the data, women are older and comparatively less likely to complete a college degree. On the other hand, men seem to have lower total cholesterol level than women.