Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 1

Turning Expectations into Edits

When we do a survey, we usually have some expectations about how the data
should look. Often, these expectations are linked to the topic of the survey.

Consider the following list of standard demographic variables:


Age Sex Marital Status
Highest Education Yearly Income Labour Force Status

1. Using these variables as a base, choose ONE of the following fictitious surveys
and briefly explain how its topic affects your expectations of the six demographic
variables:
 Ferraris and Yachts Survey: targets those who own at least one
Ferrari or yacht
 High School Students Attitudes & Grades Survey : targets all
high-school students
 Pension Survey: targets those receiving NZ Superannuation
 World War II Veteran Survey: targets NZ war veterans who served in
WWII

2. Turn these expectations into a small list of edit rules.

Remember edits rules can be fatal or query edits. Fatal edits identify errors with
certainty and query edits identify values that may be in error.
Q – Given the rules you came up with, which are fatal and which are query
edits?

Sometimes these expectations used to form edit rules are wrong. Data that does not
meet our expectations could be erroneous, or it could be trying to tell us something
such as a change in the variable we are trying to measure.
Q – How sure are you of the rules you listed?
Q – If your expectations were to be wrong, how would the data tell you this?

Now that you’ve done all this work on your expectations, think about where they have
come from. Expectations can originate from common sense, general assumptions,
or survey objectives. Some of these expectations might be more reliable than others.
Q – Can you think of any others?
Q – Where do the assumptions you’ve already made come from?

Thank you for your participation!

You might also like