ATPS lecture notes week 5, Winter 2011
*Remind them that analytical techniques are a foreign language. Counted as such in my PhD program. Just as with any language, use it or lose it. You will hear us say “I don't know”. When that happens, we will work together to find the answer.
Handout Data Dictionary ------ review-- everyone needs to create these for their surveys---- next week Tim will show you how to do something similar for interview data-
Survey monkey
BE SURE TO CLOSE YOUR SURVEY BEFORE YOU START ANALYSIS!--show
Browse responses just lets you view each respondent's responses as a whole chunk--show
crosstab is univariate only UNLESS you apply a filter.
Do crosstab first and name it and save it, --show--use for nominal and ordinal, not scale: it would go off the page and wouldn't work---would have to list out every age category for example
then apply filter and name it and save it, then you can do question by question----select "filter responses" then the questions, then "any question choice" so it will display every response you received and those where no answer was provided--- it will crosstab all questions by the one question you select---I couldn't tell a difference between the "and" vs. "or" logic
FIRST, YOU SHOULD DETERMINE WHICH WAY THE CROSS-TAB MUST BE READ
– "BY COLUMNS,” OR "BY ROWS.” The general rule could be formulated here in the
following way:
If the independent variable is in rows – read by rows;
if the independent variable is in columns – read by columns!
Note drawbacks to reporting percents vs. actual number of respondents: skew your report
Download :
do summary as PDF,
no CSV,
no advanced,
First, download the "actual" condensed results from survey monkey. This is your raw data. This way you can build your data dictionary when you download the numeric results.
Do choose "all responses collected", spreadsheet condensed, numeric value----sent to you via e-mail---don't have to recode---a little dangerous because you better make sure you know how survey monkey coded your responses----definitely will need data dictionary---have to wait for each export to complete download before you can request another
SAVE AS "ALL FILES"---name each output differently---
Excel help: http://www.audiencedialogue.net/excel1.html & Lynda.com