Nonreactive Research Neuman and Robson Ch 10

Nonreactive Research Neuman and Robson Ch. 10

Reactive vs. nonreactive research l Reactive: people being studied are aware of being studied l l l Experiments Surveys Nonreactive: Subjects are unaware they are being studied l l Unobtrusive measures Often use naturalistic settings

Varieties of nonreactive observation l Erosion measures l l – selective wear Accretion measures l – deposits of something left behind.

Strengths and Weaknesses l Strengths and advantages l l l l No subject “confounds” Can assess actual behavior rather than self-report Safety Reliability Inexpensive Good for longitudinal data Weaknesses and disadvantages l l l No control Often don’t know anything about the subjects Sample may not be representative Secondary information may have bias Need triangulation – looking at material from several different perspectives gives a more accurate view of it.

Types of nonreactive research l Some field experiments could be termed nonreactive l Naturalistic observation l Content analysis – quantitative or qualitative l Archival research l Written and audio/visual records l Secondary analysis

Content analysis l l A technique used to study written material by breaking it into meaningful units, using carefully applied rules. Use objective and systematic coding to produce a quantitative description of the observed material. Can analyze common myths Can also be used in a qualitative way l Employ semiotic techniques

Example: Erving Goffman’s Gender Advertisements (1979, 1988) Goffman combined content and semiotic analysis to look at how gender was (and still is!) portrayed in advertising. In his analysis, Goffman examined a selection of advertising images and found that women are consistently shown in subordinated positions compared to men in a variety of social situations. He also concluded that advertising both reflects and helps shape our concept of what it means to be masculine or feminine in our culture.

Erving Goffman’s Gender Advertisements l l Goffman asked the question: How is gender represented in advertising? His underlying premise is that ads are takenfor-granted pictures or displays of codified (culturally accepted) gender behaviour – the ads display ritualized behaviours.

Goffman (cont. ) l l l If we explore these codes, we can learn what it is to be “male” or “female” in our culture. Goffman believed that these codes originated in how families are structured in our society; based on the dominant – subordinate relationship between parent and child. Essentially, men treat women as they would treat subordinate males – in turn both are treated as “children”, which repeats the dominant parent – subordinate child relationship within the family.

Goffman (cont. ) l l l Note re: Goffman’s research…. Goffman showed that a fairly simple, but very cost-effective methodology like content analysis can illuminate an important theory about gender and the social world. Most theorists (ie. Leiss Kline Jhally, Waters and Ellis) believe that Goffman’s findings and the categories that he uses for his study are equally relevant today.

Example 1: The Family

Example 2: Relative Size

3. Function Ranking

4. Ritualization of Subordination

Content Analysis l What can be studied l l l Any written material Audio/visual information Useful for 3 types of research l l l Problems involving a large volume of text Research from afar or in the past Revealing themes difficult to see with casual observation.

Steps in content analysis l l l l 1. Define problem / identify the issue to be studied 2. Select the media that will be used 3. Derive coding categories 4. Sampling strategy – which sources will you use? 5. Train the coders if using 6. Code material by hand or with software 7. Analyze the data

Human vs. computer coders l Can often utilize computers l l l Internet searches Automated text search Great for extremely large sets of data Personal judgment not part of the process Cheaper and faster than humans Humans l l l Useful for coding complex concepts More flexibility Costs more time and money

Coding in a content analysis l l What gets counted? What is important for understanding themes? Structured observation – systematic observation based on careful rules Coding systems l l l Before you decide specifically on coding categories, you must specify what you are going to measure A set of rules on how to systematically observe and record content from text or images. What is the unit of analysis? l l One word One paragraph One theme One image

Characteristics of text content l l l l 1. Frequency 2. Direction 3. Intensity 4. Space Other things that could be counted: l Characters l Specific individuals Can also consider semantics – the meaning of the text l Requires interpretation l Must make judgment calls Or concepts l Crime, mental illness Themes

Manifest and Latent Content l Manifest l l Latent l l Overt or visible material – can count Symbolic content uncovered by semantic analysis – needs to be coded first (inductive process) and then counted Can use both deductive and inductive approaches to find categories (codes) for content analysis l l l Divide sample in sections Use grounded theory on a smaller portion to develop categories Use those categories on the rest of the sample.

Deductive and Inductive Category Formation l Deductive l l l Reasoning from the general to the specific Forming categories to score based on theoretical ideas. Inductive category formation l l l Reason from the specific to the general Come up with categories from data Can obtain categories by using grounded theory

Grounded theory l Theories are empirically grounded into the data. l l l 3 stages of analysis in grounded theory l l l Data collection and analysis are combined. Cycle – observe data, modify theory, observe data based on theory 1. Open coding: Find conceptual categories in the data 2. Axial coding: Look at relationship between the categories 3. Selective coding: To account for relationships, find core categories. In grounded theory, meaning derived from the data For content analysis, grounded theory can help find the appropriate codes to use. Quantitative analysis after that.

Sampling Strategy In Content Analysis l Which sources will be used? l l l Which dates will be used? What will be analyzed? l l l Depends on purpose of study, theory, etc. All of article, every 2 pages, etc. Representative sample is important Can use various sampling procedures to obtain l l l Random sampling Stratified sampling Purposive sampling – picking a sample for a particular reason.

Data Analysis in Content Analysis l Quantitative: l l l Largely depends on procedure Correlation analysis Percentages Inferential analysis Qualitative l Use semiotic analysis (developed in humanities)

Analysis of Existing Statistics and Secondary Analysis of Survey Data l l l l Also nonreactive Many sources of statistical data available Government statistics (i. e. Stats Canada, Canadian Census Data) International agencies (i. e. World Health Organization, the UN) Also many private sources Secondary analysis can be done when obtain “raw data” and do statistical analysis for your own research question Raw census data available to academic institutions

2006 and 2011 Canadian Census l 2006 Census l http: //www 12. statcan. ca/english/census 06/release/index. cfm l Population and dwelling counts Age and sex Marital status, families and households, housing Language, mobility and migration, immigration and citizenship Aboriginal peoples Labour, place of work and commuting to work, education and language Ethnic origin and visible minorities Income and earnings and shelter costs l l l l 2011 Census l http: //www 12. statcan. gc. ca/census-recensement/index-eng. cfm