Free online ZSAV to CSV converter

What the ZSAV format is

ZSAV is the compressed version of the IBM SPSS Statistics system data format. It stores tabular observations together with variable names, data types, value labels, missing-value definitions, measurement levels, and other SPSS metadata.

IBM SPSS Statistics creates ZSAV files when a dataset is saved in the compressed SPSS format. You may receive one from survey, research, statistical, or social-science software workflows. GNU PSPP and compatible Python libraries can also read many ZSAV files.

What the CSV format is

CSV stores a table as text: one record per line and one field per delimiter-separated column. Most spreadsheet applications, database systems, statistical packages, and programming languages can import it without an SPSS-specific reader.

CSV is useful for basic data exchange and inspection, but it does not preserve SPSS metadata such as variable labels, value labels, measurement levels, or user-defined missing-value rules. It also has no universal standard for delimiters, quoting, character encoding, or date representation.

How to convert ZSAV to CSV

Using IBM SPSS Statistics

The syntax method gives explicit control over whether the exported cells contain stored values or display labels:

1. Open IBM SPSS Statistics and create a syntax window.
2. Run the following command, replacing the filenames with full paths when necessary:

GET FILE='input.zsav'.
SAVE TRANSLATE OUTFILE='output.csv'
  /TYPE=CSV
  /REPLACE
  /FIELDNAMES
  /CELLS=VALUES.

/FIELDNAMES writes variable names as the first row. /CELLS=VALUES exports the stored values; use /CELLS=LABELS instead when the receiving application needs displayed value labels. The equivalent graphical workflow is File → Save As, choose a comma-delimited CSV type, select the required export options, and save the file.

Using Python

pyreadstat reads SPSS files, including many ZSAV files, and returns a pandas DataFrame:

pip install pyreadstat pandas

import pyreadstat

values, metadata = pyreadstat.read_sav("input.zsav")
values.to_csv("output.csv", index=False, encoding="utf-8")

This exports the data values and column names, not the complete SPSS metadata. Use a delimiter such as ; instead of the default comma by passing sep=";" if the destination system requires it.

Using a no-installation converter

You can convert the file for free on 101convert.com by uploading the ZSAV file, selecting CSV as the output format, starting the conversion, and downloading the result.

Quality and compatibility considerations

• Labels versus values: A categorical variable may contain numeric codes such as 1 and 2 while SPSS displays labels such as “Yes” and “No.” Choose the export mode deliberately.
• Metadata loss: Keep the original ZSAV file if value labels, missing-value definitions, variable labels, formats, or weighting information may be needed later.
• Encoding: Use UTF-8 when the data contains accented characters or non-Latin scripts, and confirm that the importing program reads the CSV as UTF-8.
• Separators and quoting: Text fields containing commas, quotation marks, or line breaks must be quoted correctly. Open the result in the target application rather than judging it from a plain-text preview alone.
• Missing values and dates: CSV has no native missing-value or date type. Check that system-missing values, user-defined missing codes, decimal separators, and date formats were interpreted correctly after import.