Unit 12 Exercises

12.1 REVIEW QUESTIONS
(A) Differentiate between measurement error and processing error.
(B) List four different types of data entry errors.
(C) List four general methods that can be used to minimize data entry problems.
(D) What is the function of an EpiData .QES file?
(E) What extension is used to identify an EpiData data files?
(F) List examples of the type of information need to document data in a code book.
(G) Why should backup copies of data files be kept off site?
(H) What is the function of an EpiData code book?
(I) What extension is used to identify a permanent SPSS data file?

The remainder of these exercises require you to create EpiData files. When creating these files, use filenames, variable names and codes exactly as specified. Backup all work to floppy.

12.2 TOX-SAMP: Cerebellar Toxicity Associated With Chemotherapy. Data represent the first 20 records from the cerebellar toxicity study by Jolson et al. (1992). Use the following variable names and codes when defining your data:

• ID
• AGE
• SEX (1 = male, 2 = female)
• MANUF (S = Smith Co., J = Jones Co.)
• DIAG ( 1 = leukemia, 2 = lymphoma)
• STAGE (1 = relapse, 2 = remission)
• TOX (1 = yes, 2 = no)
• DOSE (in gms/M²)
• SCR (serum creatinine levels in mg/dl)
• WEIGHT (weight of patient in kilograms)

Click here for the data listing.

(A) Create tox-samp.qes by clicking 1. Define data > New .QES file. Use either {curly brackets} or system defaults to create variable names exactly as specified. (Click File > Options > Create data file tab to check current system defaults.)
(B) Create tox-samp.rec from tox-samp.qes by clicking 2. Make data file.
(C) Create a check file for the data file 3. Checks > Add/revise. Use value labels as specified for SEX, MANUF, DIAG, STAGE, and TOX. Use a range restriction of 0 - 99 for the AGE variable .
(D) After clicking 4. Enter data, open tox-samp.rec and enter the data (available in the data listing) into the REC file.
(E) Create a duplicate empty REC file called tox-samp2.rec based on tox-samp.qes. Have your partner enter the data into the second REC file. Then, select Document > Validate Duplicate Files to validate the data. After identifying data errors, correct them.
(F) Generate a code book for the data file (click 5. Document > Codebook). Call this file tox-samp.not.
(G) Export the data to SPSS file tox-samp.sps and tox-samp.txt (Data in/out > Export > SPSS). Then, import the data into tox-samp.sav. Converting the .sps and .txt files into a .sav file will require the full version of SPSS--the student version does not have the syntax interpreter needed for this procedure.

12.3 HDUR: Duration of Hospital Stays. Data are from a hospital survey on utilization (Townsend et al., 1979; Rosner, 1990, p. 36). Information on the following variables are reported:

• DURation of hospitalization (days)
• AGE (years)
• SEX (1 = male, 2 = female)
• body TEMPerature (degrees Fahrenheit)
• while blood cell count (WBC: 100 / dl)
• in-hospital antibiotic use (AB: 1 = yes, 2 = no)
• whether a blood CULTure was taken (1 = yes, 2 = no)
• admitting SERVice (1 = medical, 2 = surgical)

Click here for the data listing.

(A) Create hdur.qes, hdur.chk, and tox-samp.rec. Enter the data into the data file, validate it, and then create a code book (hdur.not) for the data.
(B) Export/import these data to SPSS. Submit the files hdur.sps, hdur.txt, and hdur.sav.

12.4 WCGS: Western Collaborative Group Study (Selvin, 1991, p. 41). Data are:

• identification number (ID)
• cholesterol level (CHOL: mg/dl)
• behavior type (BEHAV: A or B)

Click here for the data listing. Create and document a reliable data file with these data.

Key to Odd Numbered Problems

Key to Even Numbered Problems (may not be posted)