[Statistical Analysis with R] Data Analysis

1. Metric(continous) variable of interest

• Is there association between diet and BMI 
• in children (ie. in the KiGGs dataset)
• and consider the effect of SES, well-being, family size and status (# of siblings, parents married/divorced etc.), lab parameters.

What to do for the investigation of this study Q?

• Create documentation/analysis/report file (R Markdown) and structure it.
• Data: Look at data dictionary, choose variables of interest (match study question to data and variables)
• Analysis plan: data preparation/check, choose appropriate statistical methods/statistics to investigate study question.
• Report structure: choose what and how to report.

Data: KiGGS03 06.RData.
Data dictionary: codeplan Kiggsneu.pdf.

Set the path and/or working directory in the beginning of the R Markdown file.

Fix R version (e.g. update at start of analysis using updateR() function in the installr package)

 

Main steps of the data analysis:

1. 1  Import dataset from an external file (e.g. xls, txt, SPSS file).

2. 2  Import check: check if dataset has been read correctly.

3. 3  Save dataset as R dataset (.Rdata), e.g. as dat raw.Rdata.

4. 4  Data check: check if data is correct/missing, and e.g. remove probands/variables or decide for imputation. Save corrected dataset as new dataset, e.g. dat corrected.Rdata.

5. 5  Transform variables, compute new variables, and/or select subset for final analysis. Save this again as new dataset, e.g. as dat final.Rdata, and use in all further steps.

6. 6  Descriptives to describe main characteristics of study sample.

7. 7  Main analyses.

8. 8  Secondary analyses.

9. 9  Sensitivity analyses.

Work on:

1. 1  Create documentation/report file (R Markdown) and structure it.

2. 2  Data: Look at variable dictionary, choose variables of interest

   (match study question to data and variables)

3. 3  Analysis plan: data preparation/check, choose appropriate statistical methods/statistics to investigate study question.

4. 4  Report structure: choose what and how to report.

5. 5  Data check: check if data is correct/missing, and e.g. remove

   probands/variables or decide for imputation. Save corrected dataset.

6. 6  Transform variables, compute new variables, and/or select subset for final analysis. Save this again as new dataset.

7. 7  Descriptives to describe main characteristics of study sample.

8. 8  Main analyses.

9. 9  Secondary, sensitivity analyses.

 

in Markdown, note in the first part of the