Research Methods In Sociology - 2206 Midterm

Scientific Method

A systematic organized series of steps that ensures maximum objectivity and consistency in researching a problem

Theory

Ways of naming, ways of conceptually ordering our senses of the world. Statements that tend to explain some aspect of social life

Casual Explanation

Direction of Influence: x>y or y>x (a cause and effect)

Parts of a Theory

Concepts, Variables, Assumptions

Observations

Taking the abstract to real life experiences.

Empirical Generalizations

Analysis of data leads to generalized statements about findings.

Direction of Theorizing

Deduction: Theory Testing

Induction: Theory Construction

Epistemology

Theory of Knowledge, how do we know what we know?

Methodology

Scientific ways of collecting information to produce knowledge

Paradigm

Ontology + Epistemology + Methodology = Worldview

Positivism

Scientific Knowledge about the real world comes from empirical observation

*Objectivity - separated from bias

Interpretivism

Our knowledge of reality, including the domain of human action, is a social constructed interpretation by human actors

*Subjectivity

Research Wheel

Theory > Hypotheses > Observations > Empirical Generalizations

Concepts

Abstract ideas expressed in words (gender, race, deviance)

Variables

Concepts that take on a range of values, quantities, or categories (years of schooling)

Key element of theory building (starting point)

Assumptions

Unobservable and untestable statements about the nature of things

Levels of Analysis

Macro-Level: Society

Meso-Level: Social Institutions

Micro-Level: Individuals and Interpersonal Relationships

Multi-Level: combination of all

Hypothesis

A tentative statement about the relation between 2 or more variables (Indv and Depv)

Independent and Dependent Variables

Independent: Causal variable

Dependent: Effect/Outcome variable

Ontology

Branch of Philosophy concerned with the study of what exists

Units of Analysis

1. Individuals (social roles, positions, relationships

2. Social Groupings (families, organizations, cities)

3. Social Artifacts (books, documents, buildings)

Ecological Fallacy

Drawing erroneous conclusions about individuals based solely on the observation of groups

Individualistic Fallacy

Drawing conclusions about groups based solely on the observation of some individuals

Quantitative Research Design

* Testing causal relationships

*Pre-determined measures for data collection

*Analysis and Finding

Qualitative Research Design

* Case Study

* Context

* Interpreting the meaning of the cases

Difference between Quantitative and Qualitative Research

Quantitative: narrow a topic into a focused question before they finalize their study design. They use research question to create a hypothesis

Qualitative: Begin with a very general research question, and as they proceed their question gradually emerges

Explanatory Variables

Independent and Dependent Variable

Extraneous Variables

All variables (control variables)

Antecedent: variable that occurs prior to IV and DV

Intervening Variable

An effect of the IV and a cause of the DV

Spurious Variable

An antecedent that affects both the IV and DV

Spurious Variable

An antecedent that affects both the IV and DV

Non-Spurious: Only affects the IV

Conceptualization

Providing a specific and tentative definition for an abstract idea or a concept

Operationalization

Linking abstract concepts to specific measurements

Measurements - assigning numbers/labels to units of analysis in order to represent conceptual properties

Two Types of Operationalizations Measures

Direct - things we can observe rather simply and rather directly

Indirect - Things that require more subtle, complex, or indirect observations

Discrete Variables

Variables that have a relatively fixed set of separate attributes

Continuous Variables

Variables that have an infinite number of values that flow along a continuum

Leves of Measurement

Nominal, Ordinal, Interval/Ratio

Nominal

Variables whose attributes only have the characteristics of being jointly exhaustive and mutually exclusive

ex. gender, religion, provinces

Mutually Exclusive Attributes

The measurement requirement that each case can be placed in only one of a variables

Exhaustive Attributes

The measurement requirement that a measure includes all possible values or attributes of a variables, so every case can be classified

Ordinal

A variable with attributes or categories you can rank order along some dimension

ex. Social class, severeness of depression

Interval/Ratio

Interval: A variable with attributes rank ordered, and have equal distances between adjacent attributes, but have no true zero value (temperature)

Ratio: A variable whose attributes are rank ordered and have equal distances between adjacent attributes, but have a true zero value (income, years of schooling)

Indices

Condense the data generated by multiple indicators into a single number/score

Scales

An ordinal, interval, or ratio measure of a variables expressed as a numerical score

Reliability and Validity

Reliability: Consistency of the measure of a variable

Validity: Goodness of fit between a definition and concept intended to measure

Face Validity

Whether the measure appears to be valid on its face

(ex. photo ID)

Content Validity

The degree to which a measure covers the range of meanings included within a concept

Construct Validity

The degree to which a measure relates to variables within a system of theoretical relationships. What it's constructed to do

Construct Validity

The degree to which a measure relates to variables within a system of theoretical relationships

External Validity

What the results mean outside the particular context of the experiment. How the study can be generalized to other situations and to other people.

External Validity

What the results mean outside the particular context of the experiment

Threats to Internal Validity: Confounding Factors

Effects of the IV cannot be separates from possible effects of extraneous variables

Threats to Internal Validity: History Effects

Something occurs during the experiment, other than the treatment, that influences the outcomes

Threats to Internal Validity: Maturation

Changes within participants that affect the results of an experiment

ex. boredom, hunger

Threats to Internal Validity: Testing Effect

Testing and retesting with influence peoples behavior (growing sensitivity)

Threats to Internal Validity: Selection Bias

Systematic differences in the composition of the control and experimental groups

Threats to Internal Validity: Mortality/Attrition

A form of selection bias when participants drop out of the study. Sometimes due to death

Threats to Internal Validity: Instrumentation

Unwanted changes in measurement

Threats to Internal Validity: Diffusion of Treatment or Contamination

Research participants in different groups communicate with each other and learn about the other's treatment

Threats to Internal Validity: Experimenter Expectancy

Searchers who indirectly communicate desired findings to the subjects

Double-blind experiment

Solution to Experimenter Expectancy

The subjects nor the person who directly deals with the subjects knows the specifics of the experiment

Threats to Internal Validity: The Hawthorne Effect

Any variability in a dependent variable that is not the direct result of variations in the treatment variable

* The tendency in experiments for people to modify their behavior because they know they are being studied, and so to distort the findings

Survey Research

A method of investigation that uses question-based or statistical surveys to collect information about how people think and act

Survey Designs

To measure multiple variables by asking people questions and then to examine the relationships among the measures

Open-ended question

A question that can be answered in any way

Closed-ended question

Only specific ways to answer the question

Mail Surveys

Advantages:

Lower Cost, Covers Wide Area, Anonymity

Disadvantages:

Low response rate, no probing, language illiteracy, coverage issues (homeless)

Online Surveys

Advantages:

Lower cost, Wide Area, Anonymity

Disadvantages: Low response rate, no probing, language illiteracy, coverage issues (no internet)

Telephone Surveys

Advantages: Wide Area, Some probing, high response rate

Disadvantages: High Cost, Interviewer Bias, Low Anonymity

Face-to-Face Interview

Advantages: High response rate, extensive probing, obtaining rich data

Disadvantages: Costly, Time Consuming, Interviewer Bias

Avoidance's when writing survey questions

1. No Jargon

2. No Ambiguity

3. No Emotional based languages

4. Prestige bais

5. Double-barreled Questions

6. Sensitive

7. Beyond respondents capacities

8. Extreme Absolutes

9. Distant future intentions

10.Double Negatives

Prestige Bias

Questions that contain an expert of authority position which can affect how people respond

Double-Barreled Question

Asking more than one item in one question.

ex. asking about gender discrimination and prejudices against minorities

Sensitive Questions

Questions that contain sensitive items that can discourage participants to respond

ex do you ever hit your wife

Extreme Absolutes

Questions that contain extreme absolutes

ex. Do you ALWAYS observe traffic signs?

5 Rules for Obtaining Usable Answers

1. Mutually Exclusive and Exhaustive

2. Open-ended questions to a minimum

3. Must Be Specified

4. Consider a Don't Know Response

5. Provide a meaningful scale

Survey Design Issues: Question Order/Sequence

General to Specific

Avoid putting sensitive Q's at the beginning

Response-Reality Gap

The difference between what study participants respond in the survey/interview and what they actually behave or act in reality

Sampling

The process of selecting a subset of cases in order to draw conclusion about the entire set

Target Population

The population to which researchers would like to generalize their results

Sampling Frame

An operational definition of the population that provides the basis for drawing a sample

ex. Phone Book, Student Registry

Coverage Error

A mismatch between the target population and sampling frame

Undercoverage- Only using phonebook

Overcoverage - Duplication (overcount)

Sampling Error (Quant Only)

The deviation of the sample from the true characteristics, traits, behaviors, qualities or figures of the entire population

Parameters

Characteristics of a population

Parameters

Characteristics of a population; Numbers that summarize data for an entire population

The Sampling Distribution

Information from the sample is linked to the population via the sampling distribution

Random Sampling/Probability Sampling

Everyone in the target population has an equal chance of being selected for a study

Theoretical Sampling/Theoretical Saturation

Qualitative approach of sampling. The point at which no new themes emerge from the data and sampling is considered complete

Transferability

The study findings' fitting outside a particular study

The findings can have meaning to another group or could applied in another context

Systematic Sampling

every kth case

Simple Random

A simple random - through generating

Stratified Random Sampling

Population is divided into strata and independent random samples are drawn from each stratum

Multi-Stage Cluster Sampling

Population is broken into smaller areas, called cluster, and a random sample of clusters is drawn

Non-probability Sampling

Cases in population are not randomly selected and they do not have an equal chance of being selected

Convenience Sampling

A non-random sample in which the researcher selects anyone he or she happens to come across

Quota Sampling

Units are selected into the sample on the basis of pre-specified characteristics, so that the total sample with have the same distribution of characteristics assumed to exist in the population being studied

Purposive Sampling

Researcher uses a wide range of methods to locate all possible cases of a highly specific and difficult-to-reach-population

ex. drug dealers, prostitutes

Snowball Sampling

The initial study participants may be asked to suggest additional people for the study

Standardized Interview

*Positivist Approach (Bias Free)

*Strictly follows a script

*Closed ended questions

- Answers are usually: multiple choice, agree/disagree, likert scale

Theoretical Background of Standardized Interview

* Stimulus-Response Model

Same questions asked in same way to everyone (bias free)