Key Psychology Research Methods and Validity Concepts

Double Blind Study

Neither participant nor experimenters working with participants know who is in the control group and who is in the experiment group.

One-Group Pretest/Posttest design

A relatively weak design; if scores on the DV change significantly from pre- to posttest, there are many potential threats to internal validity.

Confounding variable

Systematically varies with the levels of the IV.

Selection effects

Participants in independent-groups/between subjects design when participants in one level systematically differ from participants in another level.

Order/Sequence effects

Within/repeated measure design; counterbalance to eliminate.

Maturation

A change in behavior that emerges spontaneously over time.

History

Results when some external/historical events affect most members of the treatment group at the same time they're receiving treatment.

Regression (to the mean)

Statistical concept in which extremely low/extremely high performance at time 1 is likely to be less at time 2.

Attribution

Reduction in participants from pretest to posttest.

Testing

A type of order effect in which there is a change in participants' behavior simply as a result of being tested.

Instrumentation

Occurs when a measuring instrument changes over time.

Observer bias

Threat to IV caused by researchers' expectations influencing how they interpret the results.

Demand characteristics

Threat to IV when participants figure out what the research study is about and change their behavior in the expected direction.

Placebo Effects

Threat to IV that's present when people receive treatment and improve only because they believe they are receiving a valid and effective treatment.

Null Effect

No difference but may be just as significant; can also be hard to find.

Weak manipulations

0 vs. 30 minutes or 0 vs. 60 minutes; longer time has a larger effect.

Insensitive measures

Sometimes the null results occur because researchers have not operationalized the DV with enough sensitivity.

Within-group variability

Obscured the group differences.

Celing and Floor Effects

Types of insensitive measures; if the tool you're using is too easy/too hard to measure, you won't obtain effect.

Reverse Confound

Design study to include confounding variable anticipating that confound actually counteracts/reverses true effect of an IV.

Error Variance (noise)

Unsystematic variability among the members of a group in an experiment.

Quasi-Experiment

IV, also called quasi IV, serves the same purpose as IV, but researchers don't have full control.

One-Group Posttest-only

Weakest quasi experiment; there are no corresponding measures within which to compare the posttreatment results.

Posttest-Only Design with a Nonequivalent Control Group

Weak design especially as nonequivalence of the groups increase.

Pretest-Posttest Design with a Nonequivalent Control Group

Better design; preexisting differences between the treatment and control groups can be assessed.

Interrupted Time-Series Design with a Nonequivalent Control Group

Strong quasi experiment; actively seek participants to collect data before and after.

Small-N Designs

Very few participants, case study - intense study of one individual/college/region.

Stable Baseline Design

Measuring DV at different time intervals then introduce something.

Multiple baseline designs

Measuring each individual fear of dogs; vary treatment to individual.

Reversal designs

Establish the need for treatment for life; baseline, treatment, take away treatment, reintroduce.

Direct replication

Different participants/time/researchers/location.

Conceptual replication

Researchers explore the same research question, but use different procedures.

Replication plus extension

Direct replication AND add something to it like another level to the IV or a 2nd VD.

Ecological validity

The extent to which the tasks and manipulations of a study are similar to real world contexts.

P-hacking

Messing around with the numbers to obtain an effect.

Open Science

Open access, materials, data and sharing hypothesis and materials for collaboration.

Pre-registration

Before collecting any data, researchers publicly state what the studies outcome is expected to be.

Field Research

Out in the world, high external validity, give up internal.

Lab Research

Data comes to you, high internal validity, give up external.

Meta-Analysis

Way of mathematically averaging the effect sizes of all the studies that have tested the same variables.