Reference no: EM133424185
Question: You have data where you have 5,000 total participants. You have then total 6 variables, one outcome variable is depression and rest of the five variable are explanatory variables.
Outcome variables is depression (Reported Participants aged 18 years and older). Responses are recorded in data as None, Several, Majority, or ALMOST ALL.
Explanatory variables are as followed:
Physical Acitivity: Reported Participants 12 years or older. Responses is recorded in data as (YES or NO)
Age: participants are from 0 to 80. All numeric vales are recorded in data by participants.
Gender: male and female are the participants.
BMI: (Reported Participants aged 2 years or older). Participant's responses are recorded as NA. and numerical values in data
Sleep Trouble: (Reported Participants aged 16 years and older. Participant's responses are recorded as (NA. YES OR NO)
Think carefully about whether you need to limit the sample due to your outcome of interest.
For instance, if you look at the PDF with explanations of the variables from Module 1, we see that the smoking variables were only reported for participants who were 20 years or older. So you will want to sort out any cases that are younger than 20. (Filter chosen variables)?
(Depression is your outcome variable).
The depression variables were only reported for participants who were 18 or older. So you will want to sort out any cases that are younger than 18. (Filter depression variable results based on the age. No lesser than 18).
Compute summary statistics for your chosen outcome variable and at least five potential explanatory variables. ( Physical activity, Age, Gender, BMI, Sleeptrouble,)
For all categorical variables, create frequency tables in SPSS.
For all continuous variables, compute the mean, median, mode, standard deviation, quartiles, minimum, maximum, and range in SPSS
Use SPSS to compute the 95% Clopper-Pearson and Likelihood Confidence Intervals for a proportion for your outcome of interest and any other proportions of interest for the explanatory variables you chose in #2 above. If needed, you may transform a categorical variable with more than two categories into a binary variable. The Module 2 video for 6113 covers how to transform a variable in SPSS.
For instance, if you are considering BMI_WHO, you may want to turn it into a binary variable where 1 stands for BMI values that correspond to being overweight or obese and 2 stands for BMI values that are normal or underweight.
Keep in mind, that for any transformation, you will want to put the category you are most interested in first so that SPSS will model that probability first in any subsequent analysis. (This is also covered in the Module 2 Video for 6113).