Editing Psychology (section)

=== Controlled experiments ===
{{Main|Experiment}}
[[File:Flowchart of Phases of Parallel Randomized Trial - Modified from CONSORT 2010.png|thumb|Flowchart of the four phases, enrollment, intervention allocation, follow-up, and data analysis, of a parallel randomized trial of two groups modified from the [[Consolidated Standards of Reporting Trials|CONSORT 2010 Statement]]<ref name="Schulz-2010">{{Cite journal |author1=Schulz, K.F. |author2=Altman, D.G. |author3=Moher, D. |author4=for the CONSORT Group |title=CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials |journal=BMJ |volume=340 |pages=c332 |year=2010 |issue=mar23 1 |doi=10.1136/bmj.c332 |pmid=20332509 |pmc=2844940 }}</ref>]]
[[File:Milgram experiment v2.svg|right|thumb|The experimenter (E) orders the teacher (T), the subject of the experiment, to give what the latter believes are painful electric shocks to a learner (L), who is actually an actor and [[wikt:confederate|confederate]]. The subject believes that for each wrong answer, the learner was receiving actual electric shocks, though in reality there were no such punishments. Being separated from the subject, the confederate set up a tape recorder integrated with the electro-shock generator, which played pre-recorded sounds for each shock level etc.<ref name=ObedStudy>{{cite journal |last=Milgram |first=Stanley |year=1963 |title=Behavioral Study of Obedience |journal=Journal of Abnormal and Social Psychology |volume=67 |pages=371–378 |pmid=14049516 |url=http://content.apa.org/journals/abn/67/4/371 |doi=10.1037/h0040525 |issue=4 |citeseerx=10.1.1.599.92 |s2cid=18309531 |access-date=24 May 2010 |archive-date=17 July 2012 |archive-url=https://archive.today/20120717013242/http://content.apa.org/journals/abn/67/4/371 |url-status=live }} [http://www.garfield.library.upenn.edu/classics1981/A1981LC33300001.pdf Full-text PDF.] {{webarchive|url=https://web.archive.org/web/20110611105753/http://www.garfield.library.upenn.edu/classics1981/A1981LC33300001.pdf |date=11 June 2011 }}</ref>]]

A [[true experiment]] with random assignment of research participants (sometimes called subjects) to rival conditions allows researchers to make strong inferences about causal relationships. When there are large numbers of research participants, the random assignment (also called random allocation) of those participants to rival conditions ensures that the individuals in those conditions will, on average, be similar on most characteristics, including characteristics that went unmeasured. In an experiment, the researcher alters one or more variables of influence, called [[independent variable]]s, and measures resulting changes in the factors of interest, called [[dependent variable]]s. Prototypical experimental research is conducted in a laboratory with a carefully controlled environment.

A [[Quasi-experimental design|quasi-experiment]] is a situation in which different conditions are being studied, but random assignment to the different conditions is not possible. Investigators must work with preexisting groups of people. Researchers can use common sense to consider how much the nonrandom assignment threatens the study's [[Validity (logic)|validity]].<ref>Melvin M. Mark, "Program Evaluation" in Weiner (ed.), ''Handbook of Psychology'' (2003), Volume 2: ''Research Methods in Psychology''.</ref> For example, in research on the best way to affect reading achievement in the first three grades of school, school administrators may not permit educational psychologists to randomly assign children to phonics and whole language classrooms, in which case the psychologists must work with preexisting classroom assignments. Psychologists will compare the achievement of children attending phonics and whole language classes and, perhaps, statistically adjust for any initial differences in reading level.

Experimental researchers typically use a [[statistical hypothesis testing]] model which involves making predictions before conducting the experiment, then assessing how well the data collected are consistent with the predictions. These predictions are likely to originate from one or more abstract scientific [[hypotheses]] about how the phenomenon under study actually works.<ref>[[Roger E. Kirk]], "Experimental Design" in Weiner (ed.), ''Handbook of Psychology'' (2003), Volume 2: ''Research Methods in Psychology''.</ref>