Unit 2: Methods in Psychology

fact, observation

objective statement based on direct observation that reasonable observers agree is true

theory

idea designed to explain existing facts and predict new ones

hypothesis

a specific, testable prediction derived from a theory

scientific method (psychology)

objective observations about behavior (facts) lead psychologists to create conceptual models or explanations (theories), which make specific, testable predictions (hypotheses)

experiment

research method where one or more variable are manipulated (treatment is imposed) to determine their effect on others while holding all else constant; determines causation

independent variable

variable hypothesized to cause an effect on another variable

dependent variable

variable hypothesized to be affected by changes in the independent variable

within-subject design

same subjects are tested under multiple conditions of the independent variable

between-subject design

different groups of subjects experience different conditions of the independent variable

correlational study

measures relationships between variables without manipulation; cannot determine causation

descriptive study, naturalistic observation

research that observes and describes behavior without testing  relationships between variable or interference

field study

conducted in a natural environment where the researcher has little or no control over conditions

laboratory study

conducted in a controlled setting designed to manipulate variables and collect precise data

self-report method

participants describe their own behavior or mental state through questionnaires, surveys, interviews

introspection

personal observation of one’s thoughts or feelings; used by Wilhelm Wundt but criticized for subjectivity

observation method

researchers directly observe and record subjects’ behavior

hawthorne effect

alteration of behavior due to awareness of being observed

descriptive statistics

numerical methods that summarize data (mean, median, mode, standard deviation, range, correlation coefficient)

correlation coefficient (r)

statistical measure (-1 to +1) showing strength and direction of a relationship between two variables; the closer to 0 indicates no correlation, negative indicates negative correlation, positive indicates positive correlation

inferential statistics

methods to determine if results are likely due to chance or represent real effects

statistical significance

results are unlikely due to chance alone (typically p ≤ 0.05)

p-value

probability that results occurred by chance if the null hypothesis were true

factors affecting significance

size of effect, sample size, and variability of data

bias

nonrandom effects caused by extraneous factors that systematically skew results

error

random variation that can be reduced through averaging

random assignment

the likelihood that all participants in a group or a sample have an equal chance of being assigned to the different conditions within the study to counteract systematic differences

representative sample

group accurately reflecting the target population

WEIRD bias

overreliance on Western, Educated, Industrialized, Rich, and Democratic participants, limiting generalizability

reliability (replicability)

ability of a measure to produce consistent results when repeated under the same conditions

operational definition

defines variables by the procedures used to measure or observe them

interobserver reliability

degree to which multiple observers agree when measuring the same behavior

validity

the accuracy of a particular measurement device or assessment; lack of validity can lead to bias

face validity

the degree to which the material on an assessment appears on the surface to accurately measure what it intends to measure

criterion validity

the degree to which scores on a particular assessment are positively correlated with scores on another preexisting and well-established assessment tool (the criterion) for a particular skill, trait, or ability

observer-expectancy bias

researcher’s expectations influence the outcome of a study

subject-expectancy bias

participants’ expectations influence their responses or behavior

observer-blind procedure

observer does not know key aspects of the study that could bias their observations

subject-blind procedure

participants do not know key aspects of the study that could bias their responses

double-blind procedures

both researcher and subject are unaware of critical information that could bias results; achieved with use of 3rd parties

placebo

inactive treatment used to control for expectation effects

ethical principles (APA)

rules protecting participants’ rights and welfare in research

right to privacy

participants’ information and behaviors must remain confidential and anonymous

protection from harm (physical and psychological)

participants must not experience unnecessary pain or distress

use of deception, confederates

allowed only when justified and followed by debriefing

animal research ethics

animals must be well cared for, must not suffer unnecessary deprivation/pain, and must have their suffering balanced against the potential value of the knowledge gained

case study

research method that studies one unique individual (or small group of individuals with one common and unique aspect) in depth

meta-analysis

research method that analyzes what has already been studied and published to try to make new conclusions

results of correlations…

predict the possibility of causal relationships, but cannot prove them

institutional review board approval

an unbiased group that reviews research proposals to ensure researchers are abiding by all guidelines of research ethics

informed consent/assent from all participating subjects

participants maintain voluntary consent throughout duration of experiment and reserve the right to withdraw consent/from the study

debriefing of all participants

transparency of the experiment, informing participants the what, why, and the results

mean

the average of a set of numbers, found by adding all values and dividing by the number of values

median

the middle value in an ordered set of data

mode

the most frequently occurring value in a dataset

range

the difference between the highest and lowest values in a dataset

standard deviation

a measure of how much scores in a dataset vary around the mean; a small value indicates values are generally close to the average, a large value indicates values are generally spread out from the average

empirical rule: 68-95-99

in a normal distribution, about 68% of data falls within 1 std. dev. of the mean, about 95% within 2 std. dev., about 99% within 3 std. dev.

positive correlation

a relationship where two variables increase or decrease together; as x increases, y increases

negative correlation

a relationship where one variable increases as the other decreases; as x increases, y decreases

z-score

a standardized score that shows how many std. devs. a data point is from the mean; z = (data value - mean) / (std. dev)

confounding variable

an outside factor that is not the independent or dependent variable that could affect the results of the experiment, making it unclear which variable actually caused the results

control variable

any variable that researchers keep constant to prevent it from influencing the outcome of an experiment

control group

a participating group in an experiment that does not receive the experimental treatment; used for comparison to the experimental group

population

the larger group from which a sample is drawn, about which researchers will make generalizations

random selection

the selection of individuals who will be chosen from the population to participate in the study; every member of the population has an equal chance of being chosen 

random sampling

a method of choosing participants from the population randomly so that the sample is representative

content validity

the degree to which an assessment accurately measures what it claims to test

predictive validity

the degree to which a particular assessment anticipates (predicts) success on some future measure

split-half reliability

a measure of consistency which compares the results of half of a test with the results of the other half

test-retest reliability

the degree to which the same outcome is achieved on at least two occasions; comparing the results when the same individuals take the same test multiple times

alternate form reliability

the consistency of test results across different versions of the same test; compares the results of two different but equivalent versions of a test

generalizability

the extent to which study results can be applied to ore are representative of the broader population

generalizability increases when…

sample size is large, sample is randomly selected, participants are diverse and representative, the study is replicated in different settings

confederate

a person who appears to be a participant in a study but is actually working with the researcher

cross-sectional study

a research design that compares different groups of people (ex. ages) at a single point in time

longitudinal study

a research design that follows the same group of individuals over an extended period to observe changes

Clever Hans

early 1900s horse believed to perform math and language tasks; later shown his answers were guided by subtle, unintentional cues from observers, not true understanding

lessons from Clever Hans

demonstrated the importance of controlling for observer influence, skepticism, and objective measurement in psychological research

the value of skepticism

• Skeptics seek to disprove claims, which is the logical foundation of scientific testing

• A scientific theory becomes more believable as repeated, genuine attempts to disprove it fail

• Skepticism allows ppl to look more carefully and more objectively, to notice what others had missed or chosen to ignore in their experiments

skeptics seek to disprove claims, the logical foundation of scientific testing; allows more people to look more objectively to notice what others had missed or chosen to ignore in their experiments

occam’s razor

when there are two or more explanations that are equally able to account for a phenomenon, the simpler explanation is preferred

the value of careful observations under controlled conditions

to test hypotheses, scientists must control conditions in which they make observations to rule out alternative explanations

the problem of observer-expectancy effects

science is carried out by people who come to research with certain expectations; observers may unintentionally communicate to subjects their expectations, and the subject may respond by doing what the researcher expects them to do