Instructions on data files

The CIS Database is continuously working on preparing study documentation for dissemination.

Newly created studies are available on the institutional website (www.cis.es) with complete documentation once the relevant technical tasks of data coding, cleaning, validation, and anonymization, in addition to information cataloging, are completed.

In the case of older studies , those available in the Institution's archives prior to 2000, their data files are reviewed and prepared in descending chronological order by year of study, or by thematic collections, so that their documentation is also fully available on the website. The CIS is also constantly working on modernizing and expanding the formats of its data files, advancing the principle of technological neutrality and facilitating their management and understanding by the research community and others who wish to work with them.

Access to data files

To download the CIS data files, you must first complete a form. The information collected is solely for statistical processing by the institution or to notify the user, if applicable, of any changes made to the data in the requested study.

Files can only be downloaded individually, study by study; bulk downloads are not possible.

Data file format

The microdata files (data matrices) from the CIS surveys, along with the study documentation, are downloaded in a compressed file (MDxxxx.zip), which must be decompressed.

This file contains the following files, which are necessary for the exploitation of the study:

• Documentation:
  • Questionnaire (questionnairexxxx). In PDF format
  • Technical sheet (Ftxxxx). In PDF format
  • Codebook (codexxxx). In PDF format
  • Cards (cardsxxxx). In PDF format (only for surveys conducted via in-person interview)
• Anonymized microdata file (DAxxxx) in ASCII format. (See the sections 'Data file format - ASCII' and 'Annex I')
• Microdata file reading syntax program for working with SPSS (ESxxxx). In ASCII format. (See the section 'SPSS Syntax File Format' and 'Annex I')
• Other types of files that may also contain:
  • Data file in SPS format (xxxx.sav). Available for studies whose data have been prepared since 2018 (includes studies from any date).
  • Data file in CSV format, with and without labels (xxxx_etiq.csv, xxxx_num.csv). Available for studies whose data have been prepared since 2023 (includes studies from any date).
  • Microdata file reading syntax program for working with SAS (Sasxxxx). In ASCII format. Only available for studies conducted from January 2009 to November 2013 (See the 'SAS Syntax File Format' section).

When the study has several documents of the same type (for example, more than one questionnaire, more than one fact sheet, etc.), these are usually included in a compressed file. Microdata, depending on their size in bytes, can also be compressed.

Data file format - ASCII

CIS data files are provided in ASCII format (text format) and can be processed with any statistical program such as SPSS, SAS, etc.

A microdata file appears as a matrix of numbers, where the rows or records correspond to the interviewees and the columns correspond to the responses given to each variable. The columns assigned to each variable can be found in the codebook or in the questionnaire, if it was implemented on paper (PAPI), where they are indicated by a number in parentheses to the right of the response categories.

Previously, the number of columns in the matrix was limited to 80, so as many rows as necessary were used to complete the information for each interviewee. Each row of 80 columns corresponds to a RECORD CARD. The change from one card to another is indicated on the questionnaire. Starting with study no. 2083 (February 1994), the CIS questionnaires are no longer structured by cards, and the information from each interview is recorded on a single line. Both formats (with or without RECORDS) coexist in the microdata files provided by the CIS.

The microdata are anonymized to prevent individual identification of the interviewees. The variables affected in this process depend on the specific study, but are generally: interviewer, section, district, and municipalities with fewer than or equal to 100,000 inhabitants (except provincial or autonomous community capitals).

Data containing weighting variables (WEIGHT) will have a comma (,) as the decimal separator. This is taken into account when working on computers with different regional settings. Furthermore, a single study may have more than one weighting variable.

To use this file format, please read the instructions in the Appendix.

Data file format - SAV

The .sav file extension is the data file for the IBM SPSS statistical analysis program. The data is already processed and ready to be worked with directly from the application.

Data containing weighting variables (WEIGHT) will have a comma (,) as the decimal separator. This is taken into account when working on computers with different regional settings. Furthermore, a single study may have more than one weighting variable.

In the event that the data of a study require balancing, they will be provided weighted by the variable that contains the weighting coefficients of the main sample, usually called WEIGHT.

Data file format - CSV

CSV stands for "Comma Separated Values." It's an open format.

These text files look like tables where the rows correspond to the people interviewed and the columns correspond to the variables.

The most common way to read these files is to use the Excel program.

They are provided in 2 versions:

• with variable name labels and response categories ( xxxx_etiq.csv )
• unlabeled, i.e. numeric ( xxxx_num.csv )

Data containing weighting variables (WEIGHT) will have a comma (,) as the decimal separator. This is taken into account when working on computers with different regional settings. Furthermore, a single study may have more than one weighting variable.

Syntax File Format - SPSS

The syntax file is presented in ASCII format, but is designed to work with the SPSS statistical processing program.

It contains, at least, the minimum commands necessary to correctly read the microdata: the definition of variables (DATA LIST) and the weighting (WEIGHT) when the sample design requires it.

Other SPSS commands may not be included, such as variable labels (VARIABLE LABELS) and response categories (VALUE LABELS), and commands such as RECODE, MISSING VALUES, FREQUENCIES, etc. Occasionally, variable labels (VARIABLE LABELS) may appear only for multicomponent variables (multivariate, multiresponse, and multidichotomous). Newer or recently processed studies contain the full syntax.

Weighting coefficients are usually detailed in the data sheet, although sometimes they are only included in the microdata file. If a study requires more than one weighting variable (for example, PESO and PESOCCAA), only the main one, or the one that affects the overall sample, is left active.

For surveys conducted between 1992 and 1997, the syntax program may have been written in older versions of SPSS, specifically UNIX, which handles the syntax somewhat differently than newer versions. In these cases, the statements above the DATA LIST should be eliminated, or at least not executed, and command terminators, which can be a period (.) or a blank line, should be added to separate the statements.

To use this file format, please read the instructions in the Appendix.

Syntax File Format - SAS

The syntax file is presented in ASCII format, but is designed to work with the SAS statistical processing program.

At a minimum, it contains the statements necessary for correct reading of the microdata. Other statements that appear in the program are those that label the variables and response categories.

This form is no longer available and is only available for some studies conducted from January 2009 to November 2013.