Quantitative and Qualitative Research Methods in Social Sciences, Summaries of Sociology

Download Quantitative and Qualitative Research Methods in Social Sciences and more Summaries Sociology in PDF only on Docsity!

Methodological Approaches in

Social Research

Self-administered Questionnaires

Self-administered questionnaires are a type of survey where respondents are asked to complete the questionnaires themselves, without the direct involvement of an interviewer.

Questionnaire Response Rates

Questionnaire response rates refer to the number of people participating in a survey divided by the number selected in the sample. The ideal response rate is considered to be higher than 70%.

Survey Research Purposes

Survey research can serve three main purposes:

Exploratory

Surveys can be used to explore new topics or gain a better understanding of a phenomenon.

Descriptive

Surveys can be used to describe the characteristics of a population or a phenomenon.

Explanatory

Surveys can be used to examine relationships between variables and explain why certain phenomena occur.

Survey Units of Analysis

The primary unit of analysis in survey research is the respondent, the individual who provides the survey data.

Social Desirability Bias

Social desirability bias refers to the tendency of people to answer survey questions in a way that makes them look good, rather than providing truthful responses.

Contingency Questions

Contingency questions are survey questions that are intended for only some respondents, based on their responses to previous questions.

Matrix Questions

Matrix questions are a type of survey question that presents respondents with a series of statements or items and asks them to indicate their level of agreement or disagreement, often using a Likert-type scale (e.g., strongly agree, agree, neutral, disagree, strongly disagree).

Telephone Interviews

Telephone interviews are a method of survey data collection where the interviewer conducts the survey over the phone. The main advantages are the high household coverage (95.5% of households have phones), while the main disadvantages are the exclusion of unlisted phone numbers and cell phones.

Random Digital Dialing

Random digital dialing is a polling method where respondents are selected at random from a list of 10-digit phone numbers, with every effort made to avoid bias in the construction of the sample.

Computer-Assisted Telephone Interviewing (CATI)

CATI is an integrated telephone and computer system where the interviewer reads the questions from a computer screen and enters the respondent's answers directly into the computer program.

Strengths and Weaknesses of Surveys

Strengths

Ability to describe large populations Flexibility in terms of question types and topics Standardized questions

Weaknesses

"Round pegs in square holes" - surveys may not capture the full context of social life Artificial nature of the survey setting Weak on validity, as respondents may not provide truthful or accurate responses

• • • • • •

Ethics of Survey Research

Surveys often ask for private information, which must be kept confidential. Additionally, some survey questions may cause psychological discomfort for respondents.

Experiments

Experiments involve researchers selecting groups of subjects, doing something to them, and observing the effect.

Topics Appropriate for Experiments

Hypothesis testing Exploratory topics (not as much descriptive) Quantitative research Small group interaction

Independent and Dependent Variables

The independent variable is the stimulus or intervention. The dependent variable is the effect or outcome.

Pretesting and Post-testing

Pretesting is the initial measure of the dependent variable before the intervention. Post-testing is the measurement of the dependent variable after the intervention.

Experimental and Control Groups

The experimental group receives the intervention. The control group does not receive the intervention and should resemble the experimental group in all other respects.

Double-Blind Experiments

In a double-blind experiment, neither the subjects nor the experimenters know which group is the experimental or control group.

Probability Sampling

Probability sampling is a method used by pollsters to select a representative sample, where every individual in the population has an equal probability of being selected as a respondent.

• • • • • • • • • •

Random Selection

Random selection refers to the selection, assignment, or arrangement of elements by chance.

Matching

Matching is a procedure in experiments where pairs of subjects are matched on the basis of their similarities on one or more variables, and one member of the pair is assigned to the experimental group and the other to the control group.

Pre-experimental Research Designs

One Shot Case Study

A pre-experimental design that measures a single group at one point in time after they receive an intervention, with the most disadvantages out of all types of research designs.

One-Group Pre-Test Post-Test Design

A pre-experimental design that includes a pre-test for the experimental group but lacks a control group.

Static-Group Comparison

A pre-experimental design that includes both experimental and control groups but no pre-test.

Internal and External Validity

Internal Validity

The extent to which an experiment shows convincingly that changes in behavior are a function of the independent variable and not the result of uncontrolled or unknown variables. Internal validity measures what the experiment is supposed to measure.

External Validity

The extent to which we can generalize the findings of an experiment to real- world settings.

Factorial Designs

Factorial designs are analyses that involve multiple independent variables.

Natural Experiments

Natural experiments are naturally occurring events or phenomena that have somewhat different conditions that can be compared with almost as much rigor as in experiments where the investigator manipulates the conditions.

Strengths and Weaknesses of Experiments

Strengths

Isolation of the experimental variable's impact over time Ability to replicate the experiment Scientific rigor

Weaknesses

Artificiality of laboratory settings

Ethics of Experiments

Experiments typically involve deceiving subjects, and their intrusive nature opens the possibility of causing damage to participants.

Field Research

Field research involves the systematic observation of people in their natural surroundings.

Elements Appropriate for Field Research

Field research can examine practices, episodes, encounters, roles/types, personal relationships, cliques, organizations, habitats, subcultures, and other elements of social life.

Reactivity

The problem of social research subjects potentially reacting to being studied, so ultimately change their behavior from what it would have normally been.

Paradigms of Field Research

Naturalism

Based on the assumption that an objective social reality exists and can be observed/reported accurately.

•

•

•

•

Ethnography

A report on social life that is focused on detailed descriptions rather than explanations.

Ethnomethodology

Focuses on the discovery of implicit, usually unspoken assumptions and agreements (for example, breaching experiments); studying technique that deliberately disrupt social norms and observe the individuals attempt to restore normancy.

Grounded Theory

Social theory that is rooted in observation of specific, concrete details; focused on an inductive approach that attempts to generate a theory from constant comparing of unfolding observations.

Case Studies/Extended Case Method

A technique in which case study observations are used to discover flaws and to improve existing social theories.

Institutional Ethnography

A research technique in which the personal experiences of individuals are used to reveal power relationships and other characteristics of the institutions within which they operate.

Participatory Action Research

An approach to social research in which the people being studied are given control over the purpose and procedures of the research.

Focus Groups

Directed talks with a representative sample of a community, and do not involve direct conversations with individual community members.

Qualitative Interview

Contrasted with survey interviewing, the qualitative interview is based on a set of topics to be discussed in depth rather than based on the use of standardized questions.

Strengths and Weaknesses of Field Research

Pros: Finds subtle nuances, greater validity than surveys and experiences. Cons: No appropriate statistical analyses, potential problems with reliability.

Content Analysis

Study of recorded human communication (books, websites, paintings, laws, etc.).

Strengths and Weaknesses of Content

Analysis

Pros: Permits the study of processes over time, research has little effect on subjects, reliability. Cons: Limited recorded communication, validity.

Big Data

A collection of data sets so large and complex that it becomes difficult to process using on-hand database management tools or traditional data processing applications (ex. NSA, Google, Amazon, etc.).

Secondary Analysis

A research method in which researchers use existing material and analyze data that were originally collected by others. Potential cost in validity if this is used.

Analysis of Existing Statistics

Type of unobtrusive research that relies on data gathered earlier by someone else for some other purpose (different from secondary analysis); uses big data.

Evaluation Research

Research undertaken for the purpose of determining the impact of some social intervention, such as a program aimed at solving a social problem (ex. consequences of liberalized marijuana laws).

Assessment Studies

Studies that aim to determine the existence and extent of problems, typically among a segment of the population.

Cost-Benefit Studies

Studies that determine whether the cost of a program can be justified by its expense.

Monitoring Studies

Studies that provide a steady flow of information about something of interest, such as crime rates or outbreak of an epidemic.

Program Evaluation Studies

The determination of whether a social intervention is producing the intended result (also known as Outcome Assessment Studies).

Types of Evaluation Research Designs

Experimental

Just like in chapter 8.

Quasi-Experimental

Non-rigorous, lack pre/post testing and control groups.

Time Series Design

Measurement over time.

Nonequivalent Control Groups

No random assignment, otherwise similar.

Multiple Time Series

Multiple sets of data collected over time.

Logistical Problems with Evaluation Research

Getting subjects to do what they're supposed to do, because research occurs within the context of real life.

Social Indicators

Measurements that reflect the quality or nature of social life, often monitored to determine the nature of social change in a society.

Qualitative Research

Exploratory, in-depth research involving flexible, open-ended questions; includes: interviews, observation and focus groups. Used for discovering patterns in frequencies, magnitudes, structures, processes.

Cross-case Analysis

Involves examination of more than one case, either variable oriented or case oriented.

Grounded Theory Method

Theories are generated solely from an examination of data rather than being derived deductively.

Constant Comparative Method

A component of the GTM in which observations are compared with one another and with the evolving inductive theory. The use of comparisons at every stage of analysis; central to the methodological approach of grounded theory.

Semiotics

The study of signs.

Conversation Analysis

The empirical study of conversations, employing techniques drawn from ethnomethodology. Conversation analysis examines details of naturally occurring conversations to reveal the organizational principles of talk and its role in the production and reproduction of social order.

Concept Mapping

A visual means of exploring connections between a subject and related ideas and identify, graphically display, and link key concepts. Way to organize.

Coding

The transfer of words into numbers; data is transformed into a standardized form suitable for machine processing and analysis.

Manifest Content

The concrete terms contained in a communication.

Latent Content

The underlying meaning of communication.

Open Coding

The grouping of qualitative data into categories that seem logical.

Axial Coding

Analysis of categories and labels after completion of open coding.

Selective Coding

In GTM, this builds on the results of open coding and axial coding to identify the central concept that organizes the other concepts that have been identified in a body of textual materials.

Memoing

Writing memos that become part of the data or analysis in qualitative research. May describe and define concepts, deal with methodology issues, or offer initial theoretical formulations.

Code Notes

Examples of memoing; identify the code labels and their meanings.

Theoretical Notes

Examples of memoing; attempt to attach meaning to the observations (ex: why gangs do what they do).

Operational Notes

Examples of memoing; focus on data collection circumstances.

Qualitative Data Analysis Programs

Unlike quantitative analysis, which seeks to empirically describe data and test specific hypothesis, qualitative data analysis is focused on developing meaning from data.

Codebook

A guide for the qualitative analysis that outlines individual codes with definitions, criteria for inclusion, and examples.

Central Tendency

Univariate analysis; mean, median, mode.

Dispersion

The distribution of values around some central value, often the average.

Standard Deviation

A computed measure of how much scores vary around the mean; 68% within 1, 95% within 2, 99.9% within 3.

Continuous Variable

A variable (such as age, test score, or height) that can take on a wide or infinite number of values, but form a steady progression.

Discrete Variable

A variable in which the units are in the whole numbers, or 'discrete' units (for example, number of children, number of defects).

Bivariate Table

Table that illustrates the relationship between two variables by displaying the distribution of one variable across the categories of a second variable.

Contingency Table

Displays counts, and, sometimes, percentages of individuals falling into named categories on two or more variables. The table categorizes the individuals on all variables at once, to reveal possible patterns in one variable that may be contingent on the category of the other.

Elaboration Model

The initially observed relationship between two variables is mediated by a third variable.

Test (Control) Variable

'Third' Variable; held constant to clarify further the relationship between two variables.

Partial Relationship

The relationships observed after the control variable has been taken into account.

Zero-order Relationship

Just the relationship between one IV and one DV.

Explanation

Elaboration Model: when Control variable reveals a spurious relationship between original two variables.

Replication

Elaboration Model: when Control variable reveals actual relationship between two variables.

Interpretation

Elaboration Model: when Control variable is mediating factor between relationship of two variables.

Specification

Elaboration Model: when Control variable facilitates relationship between one group of zero-order relationships, but not others.

Suppressor

Elaboration Model: Control variable prevents genuine relationship from appearing at zero-order.

Distorter

Elaboration Model: Control variable reverses direction of zero-order relationship.

Ex Post Facto Hypothesis

Hypothesis created after confirming data have already been collected (it is meaningless--after the fact).

Descriptive Statistics

Computations describing either the characteristics of a sample or the relationship among variables.

Data Reduction

From unmanageable details to manageable summaries.

Proportional Reduction Error

An approach for gauging the strength of a relationship. A PRE statistic of 0 indicates that knowledge of the independent variable does not provide any help in predicting the dependent variable, and a PRE statistic of 1 indicates that the two variables are perfectly linked.

Nominal Variables

Categorical, no order (ex. gender).

Ordinal Variables

Categorical, but has order (ex. low, medium, high).

Interval Variables

Allows measured items to be ordered by rank, as well as quanitifed and compared to other items based on the differences between them.

Ratio Variables

This type of variables have order, equal intervals, and a real zero. Ex. Age.

Pearson's Product-moment Correlation

The type of correlation that summarizes linear relationships; measures strength of a linear association (measured in r).

Regression Analysis

Which statistical test is used to predict an individual's score on Y given her obtained score on X.

Linear Regression

The process that finds the best-fitting line to model a set of data; a straight line that best describes the relationship between two variables.

Multiple Regression

Linear regression of two or more independent variables on one dependent variable.

Partial Regression Analysis

Effects of one or more variables are held constant, similar to logic of the elaboration model.

Curvilinear Regression Analysis

Allows relationships among variables to be expressed with curved geometric lines instead of straight ones.

Regression Line

Y' = a + bx.

Inferential Statistics

Procedures used to draw conclusions about larger populations from small samples of data.

Nonsampling Error

All error other than sampling error; also called measurement error.

Levels of Significance

In the context of tests of statistical significance, the degree of likelihood that an observed, empirical relationship could be attributed to sampling error.

Chi Square

Is used with categorical data.

t-Test

Measure for judging statistical significance of group means. Value of t will increase with the size of differences between means, value of t will be larger when the size of the sample increases, and value of t will be larger when variations of values within each group are smaller.

Type One Error

Where the researcher rejects the null hypothesis when you should have accepted it.

Path Analysis

Path analysis is a straightforward extension of multiple regression. Its aim is to provide estimates of the magnitude and significance of hypothesised causal connections between sets of variables.

Factor Analysis

Statistical procedure designed to discover the independent elements (factors) in any set of data.

Analysis of Variance

Analysis of variance (ANOVA) is a collection of statistical models used to analyze the differences among group means and their associated procedures (such as 'variation' among and between groups).

Discriminant Analysis

Discriminant function analysis is used to determine which variables discriminate between two or more naturally occurring groups. For example, an educational researcher may want to investigate which variables discriminate between high school graduates who decide (1) to go to college, (2) to attend a trade or professional school, or (3) to seek no further training or education. For that purpose the researcher could collect data on numerous variables prior to students' graduation. After graduation, most students will naturally fall into one of the three categories. Discriminant Analysis could then be used to determine which variable(s) are the best predictors of students' subsequent educational choice.

One-Way Analysis of Variance

One-Way Analysis of Variance is a bivariate analysis technique that aims to determine which groups on the independent variable differ from each other in terms of some dependent variable. This statistical method allows researchers to compare the means of two or more groups to assess whether they are significantly different from one another.

Two-Way Analysis of Variance

Two-Way Analysis of Variance is a multivariate technique that permits the simultaneous examination of more than two variables. This approach enables researchers to investigate the effects of two independent variables on a dependent variable, as well as the potential interaction between the two independent variables.

Log-Linear Models

Log-Linear Models are a data analysis technique based upon specifying models that describe the interrelationships among variables. These models are then used to compare the expected and observed table-cell frequencies, allowing researchers to understand the complex relationships between multiple categorical variables.

Odds-Ratio Analysis

Odds-Ratio Analysis is a statistical technique for expressing the relationship between variables by comparing the odds of different occurrences. This method is particularly useful in epidemiological studies, where it can be used to quantify the association between an exposure and an outcome.

Geographic Information Systems

Geographic Information Systems (GIS) are an analytic technique in which researchers map quantitative data that describe geographic units for a graphic display. This approach allows for the visualization and analysis of spatial patterns and relationships, which can provide valuable insights into various phenomena.

Time-Series Analysis

Time-Series Analysis is the analysis of a change in a variable over time. This technique is used to identify and understand patterns, trends, and relationships in data that are collected at regular intervals, such as daily, weekly, or monthly. Time-Series Analysis can be used to make predictions, identify seasonal or cyclical patterns, and detect anomalies in the data.