shitot mechkar combined

Why might a two-level independent variable fail to capture the relationship between an IV and DV?

Because the relationship may be non-linear but monotonic; only two levels cannot reveal such complex patterns.

What are the two main strategies researchers use to go beyond a simple two-level IV design?

(1) Increase the number of levels of a single IV. (2) Increase the number of IVs by using a factorial design.

Give an example reason for increasing the number of IV levels beyond two.

There may be more than two theoretically relevant conditions to compare, such as superordinate, basic, and subordinate category levels.

List two advantages of increasing the number of levels of an IV.

It allows detection of non-linear relationships and increases the design’s sensitivity (it can also help control confounding variables).

List two disadvantages of increasing the number of levels of an IV.

It requires a larger sample size and places greater cognitive load or time demands on participants, especially in within-subjects designs.

What is a factorial design?

A design that includes two or more independent variables, with every level of each variable combined with every level of the others.

When do researchers typically choose to use a factorial design?

When they wish to study the simultaneous effects of at least two IVs or expect that the effect of one IV depends on the level of another.

In a 2 × 2 factorial design, how many experimental conditions are there?

Four conditions (cells).

How is a factorial design usually denoted?

By listing the number of levels in each IV separated by × signs, e.g., 3 × 3 or 5 × 4 × 2.

Define a \"main effect\" in a factorial design.

The overall effect of one independent variable on the dependent variable, averaged across the levels of the other IV(s).

Define a \"simple main effect.\"

The effect of one IV on the DV at a single level of another IV.

Define an \"interaction effect.\"

A situation in which the effect of one IV on the DV depends on the level of another IV.

In the biased-question example, what main effect was found for question type?

Biased (misleading) questions produced more memory errors than unbiased questions.

In the same example, what main effect was found for investigator knowledge?

Investigators who knew the crime details elicited more memory errors than naïve investigators.

Describe the interaction found in the biased-question example.

Biased questions increased memory errors only when asked by a knowledgeable investigator; with a naïve investigator, question type had little effect.

After finding a significant interaction, what analytical step should follow?

Examine the simple main effects to understand the pattern within each level of the moderating variable.

What graphical pattern indicates no interaction in a 2 × 2 design?

Parallel lines for the two levels of one IV across the levels of the other IV.

What is an IV × PV design?

A factorial design combining a manipulated independent variable (IV) with a measured participant variable (PV), such as personality or demographic traits.

Name the three participant-assignment variants of factorial designs.

Independent groups (between-subjects), repeated measures (within-subjects), and mixed designs.

In an independent-groups factorial design, how are participants assigned?

Each participant is tested in only one cell of the factorial matrix (one combination of IV levels).

In a repeated-measures factorial design, how are participants assigned?

Each participant experiences all levels of at least one IV, providing data for multiple cells.

What is a mixed factorial design?

A design that includes at least one between-subjects IV and at least one within-subjects IV.

State one benefit of using a mixed design.

It controls individual differences for within-subjects factors while avoiding carry-over effects for between-subjects factors.

What are two ways to make a factorial design more complex?

Increase the number of levels within existing IVs or add additional IVs beyond two.

How many experimental conditions does a 2 × 2 × 2 design include?

Eight conditions.

What is a three-way (triple) interaction?

An effect showing that a two-way interaction between two IVs differs across the levels of a third IV.

Provide one challenge of adding more IVs or levels to a design.

The total number of conditions grows quickly, demanding more participants and more complex analyses.

What term describes the pattern where an IV has opposite effects at different levels of another IV?

A crossover interaction.

How can increasing IV levels help neutralize confounding variables?

By including intermediate levels that control potential confounds, clarifying the true IV–DV relationship.

Why does increasing the number of IV levels often require a larger sample in between-subjects designs?

Because each additional condition needs its own participant group to maintain statistical power.

What are the three main types of research objectives discussed in the lecture?

Description, prediction, and explanation (causal understanding).

What is a pilot study?

A small-scale preliminary investigation designed to test feasibility, time, cost, risk, and potential negative effects before a larger study.

Why do researchers conduct pilot studies?

To check the viability of procedures, estimate resources, identify problems, and refine hypotheses for a subsequent full-scale project.

How do exploratory studies differ from cumulative knowledge-building studies?

Exploratory studies are first steps that map unknown phenomena, whereas cumulative studies are planned to build systematically on earlier findings.

Give an example of a basic descriptive research question from the notes.

\"What is the absolute threshold of hearing?\"

What characterises descriptive research?

Observation of given phenomena in natural settings to identify systematic patterns without manipulating variables.

What is the goal of correlational research?

To describe how two or more variables covary and to provide a basis for prediction.

Why can’t correlational research establish causation?

Because it measures naturally occurring variables without manipulation, leaving alternative explanations and temporal order unclear.

What common logical fallacy leads people to infer causation from correlation?

Post hoc ergo propter hoc (assuming that because B follows A, A caused B).

What is meant by a ‘linear bias’ in correlational analysis?

The tendency to look only for linear relationships even though variables may be related in more complex ways.

Which law illustrates a curvilinear (inverted-U) relationship between variables?

The Yerkes–Dodson Law (1908) relating arousal to performance.

Define a mediator variable.

A variable that transmits the statistical relationship between an independent and a dependent variable; without it, the direct link weakens or disappears.

Define a moderator variable.

A variable that changes the strength or direction of the relationship between an independent and a dependent variable.

What is the key difference between a mediator and a moderator?

A mediator must be present for the relationship to occur, whereas a moderator only influences the size or direction of an existing relationship.

What is a spurious correlation?

An apparent statistical relationship between two variables that is actually produced by a third, unrelated variable.

Provide a classic example of a spurious correlation mentioned in the lecture.

Ice-cream sales correlate with drowning rates because both are influenced by a third variable—hot weather.

List two practical reasons for choosing correlational research over an experiment.

(1) Ethical constraints on manipulating variables (e.g., sensory deprivation), (2) interest in prediction where causal explanation is unnecessary.

What is the primary aim of experimental research?

To establish cause-and-effect relationships by manipulating an independent variable and observing its effect on a dependent variable.

Operationally, what is the independent variable (IV) in an experiment?

The manipulated cause defined by the researcher.

Operationally, what is the dependent variable (DV) in an experiment?

The measured outcome or effect resulting from manipulation of the IV.

Name the three main procedures experimental designs use to rule out alternative explanations.

(1) Manipulation and temporal ordering, (2) experimental control of extraneous variables, (3) randomization of participants to conditions.

Why is temporal ordering critical for causal inference?

To ensure that the cause (IV manipulation) precedes the effect (DV change), ruling out reverse causation.

What is experimental control?

Holding all variables constant across conditions except the IV, so observed DV differences can be attributed to the IV.

What is the purpose of random assignment in experiments?

To distribute participant characteristics equally across conditions, minimizing systematic biases from uncontrolled variables.

How does high within-group variance affect detection of between-group differences?

It makes it harder to detect true differences between groups because individual variability masks the effect of the IV.

Define ecological validity.

The extent to which laboratory findings generalize to real-world settings and behaviours.

What is the difference between internal and external validity?

Internal validity concerns whether the IV actually caused changes in the DV; external validity concerns whether findings generalize beyond the study.

Contrast lab, field, and natural experiments in terms of IV manipulation.

Lab and field experiments manipulate the IV; natural experiments observe naturally occurring IV differences without manipulation.

What advantage does using multiple methods (mixed methods) offer researchers?

Triangulation—combining strengths of different approaches to generate hypotheses, test causality, and improve generalizability.

List one strength and one weakness of experimental designs compared with correlational designs.

Strength: allow causal inference; Weakness: often artificial settings with lower external validity.

What is internal validity?

The degree to which observed DV changes can confidently be attributed to the IV rather than confounds.

What is external validity?

The degree to which study findings can be generalized to other people, settings, and times.

What is construct validity?

How well the operational definitions of variables reflect the theoretical concepts they are intended to measure.

Which famous psychologist warned about a ‘train wreck’ related to the replication crisis in 2012?

Daniel Kahneman.

Why is replication important in research?

It tests whether findings can be reproduced, strengthening confidence in their validity and generalizability.

What four elements are specified in a research design?

Number of IVs and DVs, number of measurements of the dependent variables, number of experimental/control groups, and the way participants are assigned and sampled.

Which central feature of an experiment allows researchers to establish causal direction?

Direct manipulation of the independent variable before measuring the dependent variable.

How is ‘exclusivity’ (internal validity) achieved in an experiment?

By isolating the independent variable and using an equivalent control group that differs only on that variable.

Why do experiments provide greater sensitivity and statistical power than field studies?

Because the researcher controls the intensity of the independent variable and can fine-tune its levels.

What advantage of experiments is highlighted by the ability to repeat a study with new participants under identical conditions?

Reliable replication (repeatability).

Which two main threats to internal validity endanger a one-group pretest–posttest (before–after) design without a control group?

History and maturation (plus measurement, instrumentation, regression to the mean).

What three characteristics turn a design into a ‘true experiment’?

Random assignment, a control (or comparison) group, and experimental control over variables/procedures.

What problem does the Solomon four-group design test for, and how?

Interaction between pretesting and the treatment; it compares groups with and without pretests to detect pretest effects.

Why might a posttest-only control group design be chosen over a pretest–posttest design?

It avoids pretest reactivity and interaction between the pretest and the treatment while still benefiting from randomization.

In analysis of variance, what do the terms ‘between-group variance’ and ‘within-group variance’ represent?

Between-group variance is systematic variance due to the manipulation; within-group variance is random error originating from individual differences and measurement noise.

How does random assignment reduce systematic error (bias) between groups?

It distributes extraneous variables equally across groups, making them unlikely sources of between-group differences.

Name three methods for constructing an appropriate control group.

Randomization, matching participant characteristics, and holding a potential confounding variable constant.

What are ‘order effects’ in repeated-measures designs, and give two examples.

Changes in performance caused by the sequence of conditions; examples include fatigue and practice (learning) effects.

For what purpose is a Latin Square (counterbalancing) arrangement used?

To ensure every treatment condition appears in every ordinal position and follows every other condition equally often, controlling order effects.

Placebo, empty control, and compensatory comparison groups are all examples of what experimental element?

Different types of control groups used to isolate the effect of the independent variable.

In research methods, what does the term \"validity\" mean?

The extent to which a formal measurement tool actually measures what it is intended to measure and the extent to which conclusions and actions based on that measurement are appropriate and accurate.

What are the four major types of experimental validity identified in the lecture?

Internal validity, construct validity, external validity, and statistical conclusion validity.

What central question does internal validity ask about an experiment?

Whether the observed differences in the dependent variable were caused by manipulation of the independent variable rather than by some extraneous factor, allowing a true causal inference.

Give an example of a “history” threat to internal validity.

Any event occurring between the pre-test and post-test besides the treatment—e.g., an economic recession, war, or school closure due to COVID-19—that could influence participants’ scores and mimic a treatment effect.

What is the statistical regression (regression to the mean) threat and when is it likely to appear?

It is the tendency for extreme scores on an initial measurement to move closer to the average on a second measurement when the measure is less than perfectly reliable; it often appears when groups are chosen for very high or very low initial scores.

Why is selection considered a threat to internal validity, and how does it arise?

Because outcome differences may stem from pre-existing group differences rather than the treatment; it arises when participants are non-randomly assigned, producing unequal groups from the outset.

Which experimental design practice is recommended to control threats like selection and selection-by-interaction?

Random assignment of participants (and/or of experimental conditions) to the different groups.

What does construct validity evaluate in a study?

Whether the operational definitions used truly represent the theoretical constructs and whether the observed causal link can be generalised from the measured variables to the underlying concepts.

According to the lecture, how can you distinguish between an artifact and a confound?

An artifact is an alternative explanation that can be easily isolated and manipulated independently of the main independent variable, whereas a confound is an inseparable component of the manipulation that offers an alternative theoretical explanation (e.g., experimenter or participant expectations).

Define external validity and cite one specific interaction that can threaten it.

External validity is the ability to generalise research findings across populations, settings, times and situations; one threat is the interaction of selection and treatment, where results apply only to participants who were initially motivated, skilled, or otherwise unrepresentative.

In statistical conclusion validity, what are Type I and Type II errors?

A Type I error is falsely concluding that a relationship exists when it does not (false positive), whereas a Type II error is failing to detect a real relationship (false negative).

What is the definition of measurement validity?

The extent to which a measurement tool actually measures the theoretical construct it is intended to measure.

Why do social scientists create operational definitions for theoretical concepts?

Because theoretical constructs cannot be directly observed or measured, operational definitions translate them into observable indicators so they can be studied empirically.

Which self-report questionnaire is commonly used as an operational measure of depression in research?

The Beck Depression Inventory (BDI).

What does face validity evaluate in a measurement instrument?

Whether, on its surface, the instrument appears to measure the intended variable.

Who typically judges face validity?

Experts, judges, or the participants themselves based on logic and common sense rather than statistics.

Does establishing face validity require statistical analysis?

No. It is based on subjective impressions and logical assessment.

Can an instrument possess true validity even when face validity is low? Give an example.

Yes; for example, the Rorschach test or eye-movement measures during sleep may be valid despite lacking obvious face validity.

What is reactivity in the context of measurement?

A situation in which the act of measuring changes the behavior or responses being measured.

How can high face validity increase reactivity?

When participants easily see what is being measured, their awareness can alter their behavior and thus bias the results.