CH 7. Nonexperimental Research

Nonexperimental Research

research that lacks either manipulation of an IV, random assignment, or both

Use nonexperimental research when there is a research question about

a single variable

Use nonexperimental research when interested in a studying a

non-causal relationship between two or more variables

Use nonexperimental research when the

IV cannot be manipulated or participants can’t be randomly assigned to conditions

Use nonexperimental research when the research question is

either broad and exploratory or about what it is like to have a very specific experience

Three categories of nonexperimental research

1. Single-variable

2. Correlational & Quasi-Experimental

3. Qualitative

Correlational Research

2 variables are measured (neither manipulated) and the statistical relationship between them is assessed

Variables in a correlational research can be

quantitative or categorical

Use a correlational research rather than an experiment when the statistical relationship is not thought to be a

causal one

Use a correlational research rather than an experiment when the statistical relationship is thought to be causal but

can’t manipulate the IV

Correlational Research can show (PROS)

whether two or more variables are related

Correlational Research allows for (PROS)

general predictions

Correlational Research is often (CON)

misinterpreted

Correlational Research can’t determine (CON)

causal relationships

Correlational research has the (CON)

third variable problem

Some approaches to data collection are strongly associated with

correlational research

Data collections methods (2)

1. Observational techniques

2. Archival data

Naturalistic Observation Designs

behavior of interest is observed in the environment in which it typicall occurs

Naturalistic observation often

involves coding

coding

target behaviors are specified ahead of time and observers watch for and record those specific behaviors

Undisguised observation

participants are aware that the researcher is observing their behavior

2 approaches to observation

1. nonparticipant

2. participant

Disguised observation

participants are unaware that they are being observed

Hiding yourself is a form of

nonparticipant observation

Blend in is a form of

participant observation

you can use technology to

observe

naturalist observation has a (PROS)

realistic stetting (high ecological validity)

Naturalist observation can give you an idea of (PROS)

baseline behavior (before any manipulation)

Naturalist observation is good when you don’t know much about the (PROS)

phenomenon

Naturalistic observation cant (CONS)

establish causality

Naturalist observation often involves (CONS)

subjective judgement (vulnerable to expectancy effects)

Naturalist observation concentrates on a small sample and gives details but may not (CONS)

be generalizable

Naturalist Observation can be (CONS)

costly & time-consuming

Systematic/Laboratory observation is usually less (PROS)

costly & time-consuming

Systematic/Laboratory Observation is contrived which offers (PROS)

the researcher more control

Systematic/Laboratory Observation is subjected to (CONS)

expectancy effects

Systematic/Laboratory Observation has a greater possibility of (CONS)

subject reactivity

Systematic/Laboratory Observation is less (CONS)

flexible

Archival Data

existing data that were collected for some other purpose

Archival Data often

involves content analysis

Content analysis

family of techniques for analyzing archival data that generally involves identifying specific words, phrases, ideas, or other content in the data and then counting or summarizing their occurrence in other quantitative ways

in archival studies _____ is less of a problem (PROS)

reactivity

Archival studies are (PROS)

less time-consuming

Archival studies are prone to (CONS)

selection bias

The conclusions of archival studies are based on (CONS)

data collected by someone else (how can we be sure that the data is reliable or valid)

Quasi-Experimental Research

involves the manipulation of an IV but lacks random assignment of particpants to onditions or orders of conditions

Quasi-Experimental Research is often used in

field settings

Quasi-Experimental research in field settings is used to test the

effectiveness of a treatment

Quasi-experimental studies takes advantages of (PROS)

subject variables

Quasi-experimental studies allows for some

generalization

Quasi-experimental studies are good for

longitudinal studies

Quasi-experimental studies conclusions are open to

alternative explanations

examples of alternative explanations

-history

-maturation

-regression to the mean

-spontaneous remission

Quasi-experimental studies lack

randomization

Types of Quasi-Experimental Designs

1. Pretest/Posttest Design

2. Nonequivalent Groups

3. Interrupted Time Series Design

Types of Pretest/Posttest Design

1. single group

2. nonequivalent control groups (combination design)

Quantitative research has a research question that is

focused

Quantitative research data collection is

small amounts from large samples

Quantitative data analysis is

descriptive and inferential stats

Qualitative research has a research question that is

broad

Qualitative research data collection is

large amounts from small samples

Qualitative research data analysis is

nonstatistical techniques

Qualitative research can help (PROS)

generate new and interesting research questions and hypotheses

Qualitative research can provide (PROS)

rich and detailed descriptions of human behavior in real-world contexts

Qualitative research has issues with (CONS)

objectivity, reliability, validity, and generalizability

Quantitative research can provide (PROS)

precise answers to specific research questions

Quantitative research can lead to (PROS)

general conclusions about human behavior

Quantitative research is not good at (CONS)

at discovering new research questions

Quantitative Research does not provide (CONS)

detailed descriptions of people’s behavior and experiences

Interviews can be

structured, semi-structure, or unstructured depending on how well specified the sequences of questions or prompts is

Interviews can

have focus groups

Focus group

interview conducted with a small group of particpants at the same time

Data collection and analysis in qualitative research includes

-interviews

-participant observation

Researchers start with the data and then

develop a theory or interpretation that is grounded in those data

Steps in data analysis of qualitative research

1.Identify ideas that are repeated through the data

2. Organize these ideas int o a smaller number of broader themes

3. Write a theoretical narrative

Theoretical narrative

interpretation of the broad themes that emerged from the data, usually supported by many direct quotations or examples from the data

Mixed-Methods Research

uses both qualitative (hypothesis generation) and quantitative (hypothesis testing) methods

Triangulation

using multiple quantitative and qualitative methods in mixed-methods research to study the same topic with the goal of converging on a simple interpretation