Week 6 - Non-experimental designs

Flashcards covering core distinctions, goals, correlational research, observational research methods, survey research design, and non-experimental statistics from the lecture notes.

What is non-experimental research?

Research that lacks the manipulation of an independent variable, where researchers simply measure variables as they naturally occur.

What are the primary goals of non-experimental research?

To describe phenomena and to predict relationships between variables.

What distinguishes an experimental study from a correlational study?

How the study is conducted, specifically through manipulation (experimental) versus measurement (correlational).

How is internal validity generally rated in non-experimental research?

Non-experimental research is generally lowest in internal validity.

What type of validity is often higher in correlational research compared to experimental research, and why?

External validity, because the results are more likely to reflect real-world relationships.

What is correlational research?

A type of non-experimental research where the researcher measures two variables (binary or continuous) and assesses the statistical relationship (correlation) between them.

Why is the principle 'Correlation Does Not Imply Causation' crucial in correlational studies?

Because correlational studies do not use manipulation or control.

Define the 'Directionality Problem' in correlational research.

When two variables (X and Y) are statistically related, but it is unclear whether X causes Y or Y causes X.

What is the 'Third-Variable Problem'?

When two variables (X and Y) are statistically related because some third variable (Z) causes both X and Y.

What is a spurious correlation?

A correlation that is the result of a third variable.

What kind of relationship would Pearson's r be a poor measure for, often resulting in a value close to zero?

A nonlinear relationship, where points are better approximated by a curved line.

What is observational research?

Non-experimental studies where behavior is systematically observed and recorded, with the goal of obtaining a snapshot of specific characteristics.

Define 'Naturalistic Observation'.

Observing people's behavior in the environment in which it typically occurs (field research).

What is 'Reactivity' in research?

When a measure changes participants' behavior because they know they are being studied.

What is the Hawthorne Effect?

A specific type of reactivity where people act differently because they know they are being observed.

What is 'Habituation' in the context of observational research?

When participants become used to being observed over time and begin to behave naturally in the researcher’s presence.

What is Participant Observation and its rationale?

Researchers become active participants in the group or situation they are studying, assuming important information may only be accessible to an active participant.

What is 'Structured Observation'?

Careful observation of one or more specific behaviors in a structured setting (often lab or structured natural setting), with an emphasis on gathering quantitative data.

What are the two main characteristics of survey research?

1. Variables are measured using self-reports (questionnaires or interviews).

2. There is a strong preference for large random samples.

What are the five cognitive processes involved in survey responding, and why does their order matter?

1. Question Interpretation

2. Information Retrieval

3. Judgment Formation

4. Response Formatting

5. Response Editing

The order matters as unintended influences can arise from the context.

What are 'Context Effects' in survey research?

Unintended influences on respondents' answers based on the context in which the item appears (e.g., item order or response options).

What does the BRUSO Model provide guidelines for?

Writing effective questionnaire items: Brief, Relevant, Unambiguous, Specific, Objective.

When are 'Open-ended Items' useful in survey research?

When researchers do not know how participants might respond, allowing participants to answer in any way they choose (though they are harder to analyze qualitatively).

What are 'Closed-ended Items' in survey research?

Questions that provide a set of response options, making them easier to analyse quantitatively.

What two criteria must response categories meet for survey items?

They must be mutually exclusive (no overlap) and exhaustive (cover all possible responses).

What is the Likert Scale?

A scale that measures attitude using agreement/disagreement with several statements (e.g., Strongly Agree to Strongly Disagree).

What is 'Non-Response Bias' and how is it best minimised?

It occurs when survey non-responders differ systematically from survey responders in important ways, leading to inaccurate conclusions. It's best minimised by maximising the response rate.

What are 'Group-Comparison Research Questions'?

Research questions comparing pre-existing (naturally occurring) or self-selected groups, using one categorical independent variable (two groups) and one numeric dependent variable.

What are 'Correlational Research Questions'?

Research questions testing the association between two numeric variables.

What is 'Statistical Power' and how can it generally be increased?

The probability of correctly detecting an actual difference or effect. Increasing sample size generally increases power.

What are some examples of non-experimental statistics?

Correlational statistics examine relationships between variables:

• Cohen’s d, only works if you can define two groups

• Pearson’s r

Descriptive statistics examine relationships between variables :

• Mean, median, mode (central tendency)

• Range, variance, standard deviation (spread)

• Percentiles and percentile ranks

Chi-squared tests for associations in categorical data:

• Chi-squared goodness of fit

• Chi-squared test of independence