Ch 9 Non-Reactive Research

discourse analysis

(def) Qualitative methods for examining texts to explore how meaning, knowledge, and power are created and recreated in everyday experience

ideal types

(def) conceptual models composed of the essential characteristics of social phenomena

Basic SNA Concepts (10)

- nodes - ties - size - structural holes - cohesion/density - cliques - closeness centrality - path length - bridges - degree centrality

Content analysis is one of the nonreactive research methods available for studying topics in sociology. Some topics are more appropriately addressed by content analysis than by any other method of inquiry. Which of the following are good examples of content analysis? 1. Exploring violence on TV by selecting 50 shows at random and coding the scripts of every 10th episode at random for violent situations 2. Exploring violence on TV by selecting 100 shows at random, watching randomly picked 5-minute segments of those shows, and counting the number of violent acts such as hits, kicks, punches, and shootings 3. Exploring violence on TV by randomly choosing 500 research subjects from the phone book and asking them to complete a telephone survey about the degree of violence seen on a given TV show

1. Exploring violence on TV by selecting 50 shows at random and coding the scripts of every 10th episode at random for violent situations 2. Exploring violence on TV by selecting 100 shows at random, watching randomly picked 5-minute segments of those shows, and counting the number of violent acts such as hits, kicks, punches, and shootings

Which of the following precautions should you keep in mind when analyzing documents in your comparative and historical research? Check all that apply. 1. You always need to be wary of bias in your data sources. 2. Corroboration causes more trouble than it is worth. 3. You cannot trust the accuracy of records, no matter whether they are official or unofficial, primary or secondary.

1. You always need to be wary of bias in your data sources. 3. You cannot trust the accuracy of records, no matter whether they are official or unofficial, primary or secondary.

You are interested in whether oil companies are acknowledging or ignoring environmental issues such as global climate change. In the following table, identify the type of coding that matches each example. 1. You decide to code oil company press releases as to whether they use the words or phrases climate change, global warming, conservation, endangered species, greenhouse effect, or nonrenewable resources. 2. You decide to code oil company press releases as to whether, in your judgment, they are mentioning issues related to the environment.

1. manifest 2. latent

You are interested in what people were thinking and feeling during the days leading up to the anniversary of the September 11, 2001, attacks on the United States. You decide to look at the sentiments tweeted during the week before the anniversary that mention 9/11. You delete tweets that are not in English as well as those that are referencing the date 9/11 but seem to have nothing to do with the attacks, such as "SMU cheerleading tryouts 9/11, meet @ 9 AM on the main quad." Next, you use a random number generator to choose 100 tweets from each day of the week before the anniversary. You then code the content of the tweets using the following four categories: Safety and Security; Memorial Services; The Wars Resulting from 9/11; The Heroism of 9/11. In this study, the category "Safety and Security" includes any tweet expressing concern for one's safety or the safety of others, as well as describing safety precautions that are or are not being taken by oneself, others, an agency, province, or the federal government. In the following table are eight of the tweets in your sample. Code each tweet according to the themes listed. 1. Community vigil will be held at 6:30PM 2nite on the steps of city hall to honour lives lost in the 9/11 attacks 2. The best tribute to those that died on 9/11 would be to end ALL wars being fought in their name. 3. My work held a memorial to honour all who died in 9/11 attacks. Very moving. 4. Precautions stepped up as Biden discusses 9/11 Bomb Threat 5. Atlanta prepares for 911 with heightened security precautions at MARTA stations 6. Who were the heroic women of ground zero? Y don't we hear about them? 7. Wall Street staff remembers victims of 911 attacks with a min of silence and the singing of God Bless America. 8. Eagles are gonna swamp Rams Sunday! shout out to all the heros of 911. RIP

1. memorial services 2. wars resulting from 9/11 3. memorial services 4. safety and security 5. safety and security 6. heroism of 9/11 7. memorial services 8. heroism of 9/11

Characteristics of networks can vary along a number of dimensions. Match the correct terms with the descriptions below (size, cohesion, cliques, bridges) 1. Number of nodes 2. How tightly interconnected nodes are 3. Interconnected subsets 4. Members that connect isolated subsets

1. size 2. cohesion 3. cliques 4. bridges

In the following, indicate whether each form of data is an example of structured, semi-structured, or unstructured digital data 1. tweets 2. videos 3. government database 4. emails

1. tweets > unstructured 2. videos > unstructured 3. government database > structured 4. emails > semi-structured

Comparative and historical research differs substantially from other nonreactive research methods. Which of the following statements about comparative and historical research are true? Check all that apply. 1. Comparative and historical research is primarily a quantitative technique. 2. Comparative and historical research deals with topics such as social class, capitalism, religion, and revolution. 3. Comparative and historical research involves the use of historical methods by social scientists to examine social units over time and to compare them with one another.

2. Comparative and historical research deals with topics such as social class, capitalism, religion, and revolution. 3. Comparative and historical research involves the use of historical methods by social scientists to examine social units over time and to compare them with one another.

Which of the following statements are true about nonreactive research methods? 1. Nonreactive research methods require researchers to intrude on whomever or whatever they are studying. 2. Nonreactive research methods are always qualitative. 3. Nonreactive research methods allow researchers to learn about social behaviour without affecting it. 4. Nonreactive research methods can be quantitative or qualitative.

3. Nonreactive research methods allow researchers to learn about social behaviour without affecting it. 4. Nonreactive research methods can be quantitative or qualitative.

Coding is the transformation of raw data into a standardized form. Content analysis coding may attend to both manifest and latent coding. ____ content is the visible surface content, whereas ___ content is the underlying meaning of communications.

Content analysis coding may attend to both manifest and latent coding. Manifest content is the visible surface content, whereas latent content is the underlying meaning of communications.

deviant case testing

Examining outliers (negative cases) in comparison to typical cases to gain a sense of how generalized an insight is

In social network analysis, the actors or members of interest (for example, individuals, families, companies, and so on) are known as ____. The connections between these members are ____. The ____ of these ties can vary from strong to weak.

In social network analysis, the actors or members of interest (for example, individuals, families, companies, and so on) are known as nodes . The connections between these members are ties . The intensity of these ties can vary from strong to weak.

cliques

Interconnected subsets of members within a larger network

path length

More dense networks have a shorter average path length among members

___ involve(s) studying online behaviour and interactions in real time or as ___

Observational methods involve(s) studying online behaviour and interactions in real time or asdigital traces.

Content analysis

Researchers examine a class of social artifacts and are usually written documents (ex. newspaper articles)

analytic induction

Searching for general insights by systematically looking for patterns among individual cases

Visual Sociology

The study of society, culture, and social relationships through the analysis of audio-visual artifacts

Content analysis

The systematic study and interpretation of cultural products

Read the example and indicate if it is representative of a content analysis. Joe is interested in the language of campaign promises. He is particularly interested in the magnitude and feasibility of these promises. He decides to analyze promises made in pre-election campaign literature, speeches, and press releases.

This is an example of content analysis because it is an example of using speeches and other forms of written communication to conduct a study .

Unobtrusive data gathered online is not well suited to ___ description because the data online is not typically generated for ___ purposes.

Unobtrusive data gathered online is not well suited to thick description because the data online is not typically generated for research purposes. **Unobtrusive data gathered online usually only lends itself to thin description (Geertz, 1973) and lacks the contextual information required for deeper, more meaningful understanding (or thick description) on the part of the observer. This has to do with the nature of the data generated online, which is intended for technical rather than research purposes.

Extensive sources of data are available for analysis by comparative and historical researchers. Which of the following have been used by social scientists for comparative and historical research? Check all that apply. 1. Weekly newspapers and magazines published by a religious group 2. Court cases and laws 3. Letters written by immigrants to their families

all

A central product of culture is _____

artifacts

advantages of latent content

better designed for tapping the underlying meaning of communications

content analysis is essentially a _____

coding operation

products of culture are ____ and ____ to observation and research

concrete and subject

disadvantage of latent content

cost of reliability and specificity

Google search histories are an example of...

digital traces

advantages of manifest content

ease and reliability in coding

structural holes

gaps that exist in networks between subsets of members

Cohesion and density

how tightly interconnected nodes of a network are

Unobtrusive online research

involves observing our activities online, either in real time (e.g., chat rooms) or via digital traces (e.g., online purchases)

document analysis

involves studying published documents or static media placed on the Internet, often as a finished/authored product

Weaknesses of Content Analysis

it's limited to the examination of recorded communications

bridges

members who connect otherwise isolated cliques together

algorithmic confounding

number of the features programmed into online platforms based on computer algorithms that operate in accordance with social science research finding

Vertesehen

o (def) A method that seeks to understand the meaning of actions from the viewpoint of the participants o AKA "empathic understanding" o Max Weber used the term o researcher must be able to take on, mentally, the circumstances, views, and feelings of those being studied, so that the researcher can interpret their actions appropriately

secondary analysis definition

o (def) a form of research in which the data collected and processed by one researcher are reanalyzed—often for a different purpose—by another o Appropriate in cases of survey data

artifacts

o (def) concrete objects that are expressions of abstract culture o Ex. iPads, the Mona Lisa, Duck Dynasty, Fifty Shades of Grey, automobiles, blockbuster movies, sofas, clothes, food

structured data

o (def) database/spreadsheet formats amenable to traditional quantitative analysis techniques o ex. digital data created/stored by companies and governments in the process of providing services, generating profits, and administering laws

semi-structured data

o (def) have identifiers or tags that can be structured o Ex. emails and hashtags

unstructured data

o (def) in forms not amenable to traditional data analytical methods o Ex. photos, videos, tweets, or word-processor documents

Non-reactive Research:

o (def) methods of studying social behaviour without affecting it o Includes analysis of existing statistics, secondary analysis, content analysis, and historical and comparative research o Can be qualitative or quantitative o Ex. Durkehim's analysis of suicide > did nothing to affect suicides in one way or another

what is coding?

o (def) the process of transforming raw data into a standardized form o involves the logic of conceptualization and operationalization o ex. tv shows coded as violent, books coded as romantic

Advantages/disadvantages of secondary analysis

o Advantages § cheaper and faster § benefit from the work of professionals § meta-analysis is more likely and easier o disadvantages § whether the question that was asked provides a valid measure of the variable you want to analyze § have no assurance that the data collected by the original researcher will be appropriate for your research interests

4 types of nonreactive research methods

o Analysis of existing statistics (+ secondary analysis of survey data) o Content analysis o Unobtrusive online research o Historical and comparative research

Patterns of differences of suicide according to Durkheim

o Stability of suicide rates § found suicide rates to be about the same year after year § discovered that a disproportionate number of suicides occurred in summer § predominantly Protestant countries had consistently higher suicide rates than Catholic countries § level of economic and cultural development

basic stages of big data approaches

o collecting/capturing data from multiple sources o linking those data sets o storing and organizing the data o analyzing them

Strengths of Content Analysis

o economy in both time and money o no requirement for a large research staff o no special equipment needed o Allows correction of errors o permits study of processes occurring over a long time

Online data can be collected numerous ways

o email server logs o information from websites (Web scraping) o APIs (application programming interfaces)

Key characteristics of Big Data

o high volume of data produced o velocity of data production (including real-time streaming data) o great variety of production sources (messages, updates, images from social networks, readings from sensors, GPS signals from cell phones, and others)

Durkheim avoided the danger of ecological fallacy in two ways

o his general conclusions were based as much on rigorous theoretical deductions as on the empirical facts o by extensively retesting his conclusions in a variety of ways

Issues on both Analyzing Existing Statistics and Secondary Analysis

o limitations and constraints of available measures o sound judgment requires knowledge of how the measures were devised, collected, and coded o ethical approval almost never required

solutions to problems of validity in analysis of existing statistics

o logical reasoning: ex. Durkheim's suicide analysis o replication: all tests support hypothesis = weight of evidence supports validity of measure

Analyzing Existing Statistics

o must undertake social scientific inquiry through the use of official or quasi-official statistics o benefits § can provide a historical/conceptual context to help locate your original research § provides the main data for a social scientific inquiry

manifest and latent content: field researchers opt for?

opt for depth, preferring to base their judgments on a broad range of observations and information

unobtrusive approach

relies on archives (e.g., news reports, personnel files, and legal and/or official records) to reconstruct social network

dirty data

that many online big data sources are "always on" and collect data automatically

size

the number of nodes in a network

manifest and latent content: survey research opts for?

the use of standardized questionnaires > total specificity, even though the specific measures of variables may not be fully satisfactory as valid reflections of those variables

Unobtrusive data gathered online usually only lends itself to ___ and lacks the contextual information required for deeper, more meaningful understanding (or ___) on the part of the observer.

thin description and thick description

Secondary Analysis

you obtain a copy of someone else's data and undertake your own statistical analysis

Whole network designs

§ "bird's-eye" view of social networks § researchers collect all the information about all the actors in a network § enables them to measure the characteristics and relationships of each member of a network, as well as network-level characteristics such as cliques and structural holes

Historical and comparative research

§ (def) The examination of societies (or other social units) over time and in comparison with one another § comparative research is for many a qualitative method, one in which the main resources for observation and analysis are historical

ties

§ (def) links interconnecting these nodes § Varies in intensity · Strong: closer, long-standing relationships characterized by deeper levels of trust and obligation · Weak: more casual/superficial relationships with lesser levels of expectation and obligation

Nodes

§ (def) network members, are the units that are connected by the relations whose patterns we study § nodes (actors/members) that make up social networks can be individuals, companies, families, online accounts

manifest content

§ (def) the visible, surface content—of a communication; the concrete terms contained in a communication · Ex. counting the number of times the word love appears in each novel or the average number of appearances per page for erotic novels

Analysis of existing statistics (+ secondary analysis of survey data):

§ Ex. Durkheim's suicide analysis § There are great masses of data around you and are awaiting your use in the understanding of social life

differences between latent and manifest content

§ Manifest coding of materials (objective) involves the counting of specific elements, such as the word love, to determine whether and to what degree the passage should be judged "erotic" § Latent coding of materials (subjective) calls for the researcher to view the entire unit of analysis (a paragraph in this example) and make a subjective assessment regarding whether and to what degree it is "erotic

latent content

§ def) the underlying meaning of communications § Previous ex. might read an entire novel or a sample of paragraphs or pages and make an overall assessment of how erotic the novel was

degree centrality and closeness centrality

§ degree centrality: the number of ties a member has § closeness centrality: how closely tied to other members an actor is, as indicated by the average path length between them and other members

Egocentric network

§ examining networks from the perspective of individual actor § researcher is interested in the local network of individual nodes (the network of ties directly connected to an actor) rather than the larger network as a whole

Obtrusive approaches

§ survey interviews § contact diaries § ecological momentary assessment (EMA)

observational methods

§ those that study online behaviour and interactions either in real time or as digital traces § ex. · text-based linguistic/conversational content harvested from mailing lists · discussion forums · online chat software · information about other online activities · structures and processes of interactions and behaviour (such as friendship networks on Facebook)

steps of deviant case testing

· begin with an examination of the data · search your data to find all the cases that contradict the initial hypothesis · review each of the disconfirming cases > either give up the idea or see how it needs to be fixed