Psych as a Science and Research Methods

What is a hypothesis?

A tentative explanation that is subject to testing.

What is falsifiability, and who proposed it?

Karl Popper's idea that a scientific claim must be capable of being proven false/tested. It's what separates science from pseudoscience.

Why did Popper criticize Freud's theories?

Because Freud's explanations could account for opposite outcomes (e.g., either strict or permissive parenting could explain adult perfectionism), making them unfalsifiable and therefore unscientific.

What is inductive reasoning?

Drawing a general conclusion from specific observations (e.g., noticing a driver texting and running a red light, then concluding texting while driving is dangerous).

What is deductive reasoning?

Reasoning from a general principle to a specific conclusion (e.g., "all birds have feathers; a duck is a bird, so it has feathers"). Deductive reasoning can "prove," inductive reasoning cannot.

Can science "prove" something?

No

What is null-hypothesis significance testing (NHST)?

A statistical method testing the probability that observed data would occur if there were actually no relationship between the variables (the null hypothesis).

What is a null hypothesis vs. an alternative hypothesis?

Null hypothesis: the two variables are NOT related. Alternative hypothesis: the two variables ARE related.

What is a Type I error?

Concluding there IS a relationship between variables when there actually is NOT one (false positive).

What is a Type II error?

Failing to detect a relationship between variables that actually DOES exist (false negative).

What does p < .05 mean?

That if the study were repeated 100 times, this result would occur by chance fewer than 5 times

What is a scientific theory (vs. an everyday "theory")?

A comprehensive, well-supported, testable, and falsifiable framework explaining a phenomenon

What are "levels of analysis"?

The idea that one phenomenon can be explained at multiple levels simultaneously (e.g., biological, cognitive, behavioral, social/cultural)

According to Thomas Kuhn, is science purely objective?

No

Research Designs

What's the difference between correlational and experimental research?

Correlational research measures naturally occurring variables to see how they relate (cannot determine causation). Experimental research manipulates one variable to observe its effect on another (can determine causation).

What is an independent variable?

The variable the researcher manipulates/controls.

What is a dependent variable?

The variable that is measured; it "depends" on the independent variable.

Why is random assignment so important in experiments?

It ensures that, on average, groups are similar on all characteristics except the manipulated variable, allowing the researcher to attribute differences in outcomes to the independent variable itself.

What is a confound?

A factor other than the independent variable that could explain differences between groups, undermining causal conclusions.

What is the placebo effect?

When simply believing you're receiving special treatment (like a real drug) changes your behavior or perception, regardless of the treatment's actual effect.

What is participant demand?

When participants behave the way they think the experimenter wants them to behave.

What is a double-blind procedure, and why is it used?

Neither the participant nor the experimenter knows which condition/treatment the participant is receiving

What does a correlation coefficient (r) tell you?

The direction (positive or negative) and strength of the relationship between two variables. Values range from -1 to +1; the closer to ±1, the stronger the relationship.

Positive vs. negative correlation

what's the difference?

Why does "correlation does not equal causation"?

Because with a correlation alone, you can't rule out reverse causation (B causes A) or a third variable causing both A and B.

What is a quasi-experimental design?

Similar to an experiment (has IV and DV) but without random assignment

What is a longitudinal study?

A study that follows the same group of people over an extended period of time (weeks, months, years, or decades).

What are qualitative research methods? Name three.

Methods used to study hard-to-manipulate topics: participant observation (researcher embeds in a group), case studies (intensive study of one individual/context), and narrative analysis (studying people's stories/personal accounts).

Why are surveys useful, and what type of research do they typically support?

They reach large numbers of people at low cost and time investment; typically used for correlational research (though can be used experimentally too).

The Replication Crisis in Psychology

What is replication?

Repeating a study to determine whether the same findings hold up across different times, samples, and situations.

What is exact (direct) replication?

Recreating a study's methods as closely as possible to see if the same results occur.

What is conceptual replication?

Testing the same underlying hypothesis/idea using different specific methods or measures.

Roughly what percentage of studies replicated in the Open Science Collaboration (2015) project, and who led it?

About 1/3 overall replicated (ranging from 23% to 53% across journals); the project was led by Brian Nosek.

List four reasons for non-replication.

(1) Faked/falsified data, (2) small sample sizes producing chance findings, (3) findings that were only true for certain people/circumstances/time periods (not universal), (4) errors in how the replication itself was conducted.

What is priming?

The process by which exposure to one stimulus (often subtle/subconscious) makes certain related thoughts, feelings, or behaviors more accessible or likely.

Name the three solutions to the replication crisis discussed in the module.

(1) Creating outlets to disseminate replication attempts (e.g., PsychFileDrawer, Open Science Framework), (2) pursuing more systematic/programmatic research (follow-up studies rather than one-off flashy findings), (3) reforming textbook and journal practices (e.g., being cautious about single-study findings and correcting persistent inaccuracies).

Why isn't all non-replication necessarily "bad" for science?

Because original studies venture into uncertain territory