longhunt / sci_fi_survey Goto Github PK

SCIENCE FICTION COLLECTION SURVEY

This repository contains data files for a survey being conducted by
Creative Minority Productions (http://creativeminorityproductions.com)
as a community service project in conjunction with a local public 
library branch.

Unless otherwise noted all files are released under a Creative Commons
Attribution-Noncommercial license (see file LICENSE for details).

The survey has two main fpurposes: 1) To discover the relative 
importance of various factors in patrons' selection of speculative 
fiction materials. 2) To segment the speculative fiction using 
population so that the needs of each segment can be more adequately met.

Analysis is currently ongoing; this repository currently contains mostly
raw data or data in a very early stage of analysis.  I will post the 
actual report as soon as I have a good first draft.

Kevin A. Straight   7 June 2018


FILES

code_data.py           Python 2.x script used to code the raw dump from 
                       Google Forms into a numeric format usable for 
                       analysis (GPL 2.0 License)
                 
finalSurveyDump.csv    Raw data dump from Google Forms

finalDumpClean,csv     Raw data after procressing with code_data.py.  
                       Variables coded numerically, empty records 
                       removed. Invalid and foreign zip codes deleted 
                       within records.
                       
LICENSE                Full text of the CC-BY-NC license

README                 This file

survey_facimile.pdf    Print-out of the actual survey

survey_data.odt        k-means cluster analysis (Open Office 
                       spreadsheet)
                       
survey_data.por        Survey data as a PSPP files (should open in SPSS
                       or import into other stat packages, but I haven't
                       tried yet)
                       
survey_data.xls        k-means cluster analysis saved as an Excel
                       spreadsheet (native format is .ods)
                       
T-test.odt             Output of PSPP for unpaired-T tests for various
                       splits of the data, included for people who 
                       can't run PSPP or SPSS for some reason.
                       
T-test.pdf             " 

zip_code_database.csv  List of valid us zip codes.  Used by 
                       code_data.py.  This comes from USPS so I believe
                       it is in the Public Domain under the Federal 
                       Documents rule.  If I'm wrong, someone please
                       correct me.