Research with Nonreactive

Measures
Chapter 8

This multimedia product and its contents are protected under copyright law. The following are
prohibited by law:
any public performance or display, including transmission of any image over a network;
preparation of any derivative work, including the extraction, in whole or in part, of any images;
any rental, lease or lending of the program.

Analyzing Physical Evidence for

Clues about Social Life
Non-reactive research = research
techniques in which the people in the
study are unaware that someone is
gathering information or using it for
research purposes.
Four types of non-reactive research:
Physical evidence analysis
Content analysis
Existing statistics analysis
Secondary data analysis

Copyright Allyn & Bacon 2009

Analyzing Physical Evidence for

Clues about Social Life
Unobtrusive measures = non-reactive
research measures that do not intrude or
disturb a person, so they are unaware of
them.
Limitations of Physical Evidence
physical evidence measures are indirect
possible privacy violation

Copyright Allyn & Bacon 2009

Revealing the Content Buried within

Communication Messages
Content Analysis
Content analysis = A non-reactive technique
for studying communication messages.
Text = In content analysis it means anything
written, visual, or spoken in a communication
medium.
Content analysis is useful for
Large volumes of text
Topics studied at a distance
Content difficult to see with casual observation
Copyright Allyn & Bacon 2009

Revealing the Content Buried within

Communication Messages
How to Measure and Code in Content
Analysis
Coding System = In content analysis a set of
instructions or rules stating how text was
systematically measured and converted into
variables.

Copyright Allyn & Bacon 2009

Revealing the Content Buried within

Communication Messages
How to Measure and Code in Content
Analysis
What do you measure?
Direction
Frequency
Intensity
Space
Prominence

Copyright Allyn & Bacon 2009

Revealing the Content Buried within

Communication Messages
How to Measure and Code in Content
Analysis
Coding, Validity, and Reliability
Manifest coding = Content analysis coding in which
you record information about the visible, surface
content in a text.
Latent coding = Coding in content analysis in which
you look for the underlying, implicit meaning in the
content of a text.
Intercoder reliability = A measure of measurement
consistency in content analysis when you have
multiple coders.
Copyright Allyn & Bacon 2009

Revealing the Content Buried within

Communication Messages
How to Measure and Code in Content
Analysis
Coding, Validity, and Reliability
Content Analysis with Visual Material
visual material communicates indirectly
visual images often contain mixed messages and
operate at multiple levels of meaning

Copyright Allyn & Bacon 2009

Revealing the Content Buried within

Communication Messages
How to Conduct Content Analysis
Research
Step 1. Formulate a Research Question
Step 2. Identify the Text to Analyze
Step 3. Decide on Units of Analysis
Step 4. Draw a Sample

Copyright Allyn & Bacon 2009

Revealing the Content Buried within

Communication Messages
How to Conduct Content Analysis
Research (cont)
Step 5. Create a Coding System
Step 6. Construct and Refine Categories
Step 7. Code the Data onto Recording
Sheets.
Step 8. Data Analysis

Copyright Allyn & Bacon 2009

Revealing the Content Buried within

Communication Messages
Limitations of Content Analysis
content analysis cannot:
determine the truthfulness of an assertion
evaluate aesthetic qualities
interpret content significance
reveal the intentions of the texts creators
determine the influence of a message on its
receivers

Copyright Allyn & Bacon 2009

Mining Existing Statistical Sources to

Answer New Questions
Social Indicator = Any measure of social
conditions or well-being that can used be
used in policy decisions.
Locating Data
many diverse sources
Statistical Abstract of the United States

Copyright Allyn & Bacon 2009

Mining Existing Statistical Sources to

Answer New Questions
Verifying Data Quality
Existing statistics can be limited by
1.
2.
3.
4.
5.

Missing Data
Reliability
Validity
Topic Knowledge
Fallacy of Misplaced Concreteness = when
statistical information is reported in a way that
gives a false impression of its precision.

Copyright Allyn & Bacon 2009

Mining Existing Statistical Sources to

Answer New Questions
Verifying Data Quality (cont)
Existing statistics can be limited by
6. Ecological Fallacy = mistaken interpretations that
occur when you use data for a higher or bigger unit of
analysis to examine a relationship among units at a
lower or small unit of analysis

Creative Thinking About Variables of Interest

Standardization of Data
Standardization = adjusting a measure by
dividing it by a common base to make
comparisons are possible.

Copyright Allyn & Bacon 2009

Answering New Questions Using

Survey Data Collected by Others
Secondary Sources
General Social Survey = A large-scale
survey with many questions of a large
national sample of adult Americans
conducted almost every year. Data from it are
made available to researchers at low or no
cost.
Limitations of Secondary Data Sources
May lack data for your research question
Validity
Topic knowledge
Copyright Allyn & Bacon 2009

Answering New Questions Using Survey

Data Collected by Others

Copyright Allyn & Bacon 2009

Conducting Ethical NonReactive

Research
Protect peoples privacy
Protect confidentiality of data

Copyright Allyn & Bacon 2009