Multiple Indicator Cluster Surveys
Data Processing Workshop
Creating Analysis Files:
Description of Preparation Steps
MICS4 Data Processing Workshop
Secondary Data Processing Flow
Export Data from CSPRO
Import Data into SPSS
Recode Variables
Add Sample Weights, Wealth Index and GPS Data
Run Tables
MICS4 Data Processing Workshop
Exporting ‘clean data’ from CSPRO to SPSS
– By using CSPRO dictionary and final data file,
we create ASCII rectangular data files (*.dat)
and corresponding syntax files (*.sps) to read
the data into SPSS
– All located in ‘c:\mics4\spss’ directory
• c:\mics4\spss\myhh.sps
• c:\mics4\spss\myhh.dat
•…
MICS4 Data Processing Workshop
Rectangular data files [*.dat]
•
•
•
•
•
myhh.dat - [Household]
myhl.dat – [Household members]
mywm.dat – [Women]
mych.dat – [Children]
mytn.dat – [Mosquito nets]
– Only if ‘ITN’ module’ is included
• myfg.dat – [FGM]
– Only if ‘FGM’ module’ is included
• mybh.dat – [Births to women 15-49]
– Only if ‘Birth History’ module is included
• mymn.dat – [Men]
– Only if ‘Men’ questionnaire is included
MICS4 Data Processing Workshop
Rectangular data files [*.dat]
– HOUSEHOLD QUESTIONNAIRE
• myhh.dat - [Household]
• myhl.dat – [Household Members]
• mytn.dat – [Mosquito nets]
– INDIVIDUAL WOMEN QUESTIONNAIRE
• mywm.dat – [Women age 15-49]
• myfg.dat – [Female genital cutting. women age 15-49]
• mybh.dat – [Births to interview. women age 15-49]
– UNDER-5 QUESTIONNAIRE
• mych.dat – [Children]
– INDIVIDUAL MEN QUESTIONNAIRE
• mymn.dat – [Men age 15-59]
MICS4 Data Processing Workshop
Rectangular data files [*.dat]
• Rows are ‘cases’
–
–
–
–
–
Household – all (interviewed or not)
Members - living in interviewed households
Women – all (interviewed or not)
Children – all (interviewed or not)
Nets – mosquito nets in interviewed household
• Only if ‘ITN’ module’ is included
– Daughters – daughters under 15 years of age of interviewed women age 15-49
• Only if ‘FGM’ module’ is included
– Births – births to interviewed women age 15-49
• Only if ‘Birth History’ module is included
– Men – all (interviewed or not)
• Only if ‘Men’ questionnaire is included
• Columns are ‘variables’
– Cluster, Household number, Line number, Interviewer number, etc.
Screenshot
• Sample ASCII data file
Syntax files (to read rectangular data files)
•
•
•
•
•
myhh.sps
myhl.sps
mywm.sps
mych.sps
mytn.sps
– Only if ‘ITN’ module is included
• myfg.sps
– Only if ‘FGM’ module is included
• mybh.sps
– Only if ‘Birth History’ module is included
• mymn.sps
– Only if ‘Men’ questionnaire is included
•
Includes the location information (columns) of variables in rectangular data file as well
as variable labels and value labels
MICS4 Data Processing Workshop
Screenshot
• Sample Syntax file
Importing Data to SPSS
– Executing syntax files created by CSPRO
– In SPSS, run all my*.sps files (one by one)
• The my*.sps files will import the data into SPSS
and create SPSS-format data files (system files *.sav), with variable labels, value labels etc.
– At the end of each run, save your work in data
files as hh.sav, hl.sav, wm.sav, ch.sav
[tn.sav, fg.sav, bh.sav, mn.sav]
MICS4 Data Processing Workshop
Saving the Data File
• It is helpful to have the syntax file
automatically save the data file that it
creates
• To do so, add the following line to the end
of the syntax file:
save outfile = 'hh.sav'.
MICS4 Data Processing Workshop
Screenshot
• SPSS Data Editor – Data View
– Household
Screenshot
• SPSS Data Editor – Variable View
– Household
Run frequency distributions
• Open SPSS and run frequency
distributions for all of the variables in all 7
files
• Detect any unexpected codes or
inconsistencies, and if necessary, return to
CSPro and make corrections.
Next Steps
• Recoding variables
– Creating new variables by recoding existing
variables
– Adding variables to different levels
• Mother’s education
• Father’s education
MICS4 Data Processing Workshop
Customize make*.sps
• Customize the make*.sps syntax files, in
accordance with the contents of your
questionnaires
• Take out any variables and associated
syntax which are not included in your
questionnaire, or modify the variables
according to your response categories
Run make.sps
• Run make.sps to recode all additional
variables needed for the tabulations
• make.sps is provided as a batch file which
calls and runs each of the make*.sps files
in turn
• All *.sav variables will be replaced!
• Backup before you proceed to this stage
Next Steps
• Adding sample weights
– Sample weights are added from weights
spreadsheet
MICS4 Data Processing Workshop
Calculate weights
• Obtain distribution of interview results by
the unit based on which you intend to
calculate unique sample weights:
- using the weights_table.sps syntax file, if
calculation at cluster level
- using the weights_table_alt.sps syntax file, if
calculation at stratum level, e.g. regions and
area
Calculate weights cont.
• Calculate final sample weights
• using weights.xls, if calculations are
carried out at the cluster level
• weights_alt.xls, if calculations are carried
out at stratum level
Calculate weights cont.
• Once calculations in weights.xls or
weights_alt.xls have been completed,
save the “Output” workbook
– c:\mics4\weights\weights.xls
Append sample weights
• Append sample weights to the exported
SPSS files hh.sav, hl.sav, wm.sav, ch.sav
(tn.sav, fg.sav, bh.sav and mm.sav if
relevant modules included), by using the
SPSS weights_merge.sps or
weights_merge_alt.sps syntax file
– Need customization if ITN, Birth History
modules or Men questionnaire included
Next Steps
• Adding wealth index
– Wealth index calculated then added to files
MICS4 Data Processing Workshop
Customize wealth.sps
• Customize the wealth.sps syntax file that
calculates wealth index values, together
with the survey coordinator/technical
manager, according to the items included
in your questionnaire
Run and append wealth
• Run the customized SPSS syntax
wealth.sps to calculate and append
wealth index values to all of the exported
data files
Next Steps
• Adding GPS data
– Geographic location data added to files
MICS4 Data Processing Workshop
Next Steps
• Tabulation
– Tables are generated from the *.sav files
MICS4 Data Processing Workshop
Customize each syntax
• Customize all tabulation syntax files in
accordance with the contents of your
survey
• Delete the syntax files which are used to
obtain tables on topics which are not
included in your survey
Run all_tables.sps
• Once all tabulation syntax files are customized
and free of errors, customize all_tables.sps
• all_tables.sps calls and runs all tabulation
syntax files, and produces an output which
includes all tables
• If any of the tabulation syntax files have errors,
the program will stop and you will have to
correct the tabulation syntax file. Then re-run
all_tables.sps