Research Methods: Chapter 2

Is psychology a science?

YES

• All sciences try to explain the world and uncover reality

• Need for use of scientific method to make legit claims

What is the structure of the theory-data cycle? What does the cycle look like?

Theory → Research questions → Research design → Hypothesis → Preregistration of hypothesis → Data

Variables

Things that vary from person to person and place to place (weight, age, intelligence, income, mood, etc.)

Measured variables

Measured as they exist; every study has them (e.g., measuring height, weight, age, personality)

Manipulated variables

Researcher decides who gets what; only some research designs have them (e.g., randomly assigning participants drug dosages)

What are operational definitions?

Definitions of a variable as they are measured in a given study; how a variable operates in a study

No correct options, but one might be better for your research question

“Operationalizing” a variable means…

Choosing how you’re going to define and measure/manipulate your study’s variables; turning an abstract concept into something concrete and measurable

What are the cons and pros of surveys?

Pros: allows you to measure variables that cannot be measured with other methods (e.g., attitudes, memories, personality, emotions)

Cons: people might lie, not be paying attention, not remember, or not know the answer

What are the potential issues with observation?

Reactivity, observer bias, and observer effect

Reactivity

Participants act differently just because they’re being watched

Observer bias

Researcher’s expectations affect their interpretation of what they’re seeing

Observer effect

Researcher’s expectations affect how they interact with participants → changes participants’ behavior

What is the difference between a population and a sample?

• Population: the group the researchers wants to make claims about BUT they cannot study everyone in population of interest

  • Sample: smaller subset of population that we collect data from

What are sampling strategies?

Random sampling and non-random sampling

Why is random sampling better than non-random sampling?

Random sampling: MOST representative and least biased; every population member has equal chance of being selected and results are generalizable

Non-random sampling: usually based on convenience

What are the types of study designs?

Case studies and correlation studies

Case study

In-depth study of a very small sample in a highly unusual situation; provides a deep understanding of an unusual phenomenon

What are the pros and cons of case studies?

Pro: can spark unique insight and new research questions

Con: limited generalizability

Correlation study

Measuring two variables to see if they’re related; shows the strength and direction of a relationship BUT CANNOT establish cause/effect

Does correlation equal causation?

NO, do NOT conclude cause/effect based only on correlation; relationship between two variables is typically caused by a third variable

What is the correlation coefficient r?

Tells you the strength and direction of the relationship between two variables

• Ranges from -1.0 (strongest possible negative relationship) to +1.0 (strongest possible positive relationship)

• 0.0 = no correlation at all

What does it mean to have positive, negative, and no correlation?

Positive: one variable increases ↑, other increases ↑

Negative: one variable increases ↑, other decreases ↓

None: one variable increases ↑, other stays the same =

What are the steps in experimental design?

1. IV is manipulated (researcher decides who gets what; random assignment)

2. DV is measured

3. Compare groups to see if there’s a difference

What does random assignment do?

Allows you to conclude that the change in the DV was due to the IV and not some other variable

What do experiments do?

Conclude cause/effect; experiments must meet the 3 steps to make legit claim

What are confounds?

Alternative explanations for the relationship between the IV and DV; variables that interfere with the experiment’s variables

Why are confounds important in experiments?

Researchers must prevent confounds from affecting results (what-ifs)

What are the types of validity?

Construct validity (how well were the variables measured?)

External validity (how well can the results generalize beyond the study?)

Internal validity (did the study rule out alternative explanations?)

Statistical validity (do the data support the study’s claim?)

Construct validity

How well a study’s variables were defined, measured/manipulated, and operationalized

Reliability = consistency

Validity = accuracy

External validity

How well the sample represents + how well the study’s results can generalize the population of interest

• Increased by random sampling

Internal validity

Ruling out alternative explanations for a cause/effect conclusion + confounds as an explanation for results

Being able to conclude that any difference in the DV is due to the IV and not some other variable

• Increased by experimental design

• Random assignment → groups are equal at start

What is the difference between random sampling and random assignment?

Random sampling recruits participants in an unbiased way and leads to greater generalizability of results → higher EXTERNAL validity

Random assignment manipulates an IV, is an experimental feature that allows cause/effect conclusions, and concerns how participants are grouped AFTER sampling → higher INTERNAL validity