Index A, B, C, D, E, F, G, H, I, J, K, L, M, N, O, P, Q, R, S, T, U, V, W, X, Y, Z
Affirming the Antecedent: In a conditional syllogism, the second premise states that the if part of the first premise is true. This is a valid syllogism, where the conclusion is the part of the first premise following the then.
Affirming the Consequent: In a conditional syllogism, the second premise states that the then part of the first premise is true. This is an invalid syllogism.
Alpha (a) level: The probability of a Type I error. It is used with inferential statistics. If the probability of the null hypothesis (the p value) is true is less than a then the null hypothesis is rejected.
Alternative Hypothesis: The statistical hypothesis that proposese that variables are related or groups are different. The specific form(s) of the a hypothesis depends upon the inferential statistic.
Biased sample: A sample that differs systematically
from the population.
Central Limit Theorem: The theorom that describes the sampling distribution of the mean. It has three parts:
Concurrent Validity: A measure of validity were the psychological measure is correlated with current behavior.
Conditional Syllogism: A syllogism where the first premise is structured as an if..then... statement.
Confound Variable: An extraneous variable that varies along with the independent variable of an experiment, thus rendering the experiment internally invalid.
Construct: An abstract variable that must be measured indirectly, such as self esteem or heat. In psychology these constructs are thought to be features of a persons psychological make-up. They are also called intervening variables.
Convergent Validity: Use of several different measures together that seem to indicate the same result as a measure of validity.
Correlational Coefficient: A statistics that measures the strength of relationship between two or more variables. The use of this statististic does not imply what type of research design was used. It could be ether correlational or experimental. There are several types of computations of this coefficient. See Pearson's for an example.
Correlational Design: A research design that measures two or more naturally varying variables. It does not manipulate any of the variables.
Criterion Validity: The ability of the test to correlate with some external referent, often a behavior.
Critical Value: In statistics, this is the value of the statistic with a probability equal to the alpha level.
Datum (plural is Data): A single measurement. Data is plural, refers to several measurements, and should be only used with plural verbs, e.g. Data are ... not Data is....
Deception: Under certain circumstances, it is allowable to withhold some information about a study from subjects. Check http://www.apa.org/ethics/code.html#6.15 of the APA code for allow uses and conditions on deception.
Demand Characteristics: these are behaviors of subjects in studies that are modified by cues they pick of as to the nature of the study.
Deny the Antecedent: In a conditional syllogism, the second premise states that the if portion of the first premise is false. It is an invalid syllogism.
Deny the Consequent: In a conditional syllogism, the second premise states that the then portion of the first premise is false. It is a valid syllogism and the conclusion is the negation (not) of the if part of the first premise.
Dependent Variable: The variable that is measured in an experimental design.
Determinism: The idea that behavior is controlled by natural law (natural causes).
Discriminant Validity: This is the type of validity where the measure discrimates between groups as it ought to, e.g., discriminating between those with and and those without some disorder.
Double-Blind Study: A study design where neither the participant nor the researcher running the subject know the condition or treatment that that particular participant is in.
Element of a sample: A single value drawn from a population.
Emergent Properties: The theoretical position that more complex phenomena have features that are not explanable in terms of more basic aspects of nature. For example, biological organisms cannot be explained soley by understanding their chemistry. This position opposes reductionism.
Empirical: The idea that the knowledge gained is based on observation, i.e., directly observing the world and not originating from other sources such as thinking about the world or logical necessity.
Event Sampling: Records the occurance of each event that meets specified criteria.
Experimental Design: A research design where one or more variable is manipulated, other variables are controlled or balanced and a dependent variable is measured.
External Validity: A study may be generalized to the population.
Falsification, Principle of: The logical principle that it is only possible logically to prove a theory or hypothesis false. It is not possible to prove a theory or hypothesis true. It is based on the reasoning related to Affirming the Consequent.
Field Experiment: An experiment conducted in a natural situation. It has both an independent and dependent variable (s).
Histogram: This is a type of graph where a variable is plotted on the X-axis and the frequency of each value of that variable is plotted on the Y-axis. This is illustrate below. See also normal distribution.
Hypothesis: A testable prediction. see also Principle of Falsification.
Hypothetical: Some variable that is proposed to operate psychologically.
Independent Variable: The variable that is manipulated by the experimenter.
Inferential Statistic: A statistic used to generalize from the sample to the population.
Informed Consent: Subjects have the right according to APA and legal guidelines, to be informed of all relevant features of an experiment before participating. See also deception.
Internal Validity: An experiment is design to allow a causal conclusion to be made.
Inter-rater reliability: The extend to which two observers score an observed situation the same. Measured with Pearsons's correlation coefficient, usually.
Intervening Variable: See construct.
Law: A regularity observed in nature. Often refers to the description of this regularity.
Manipulation Check: an additional measurement made to determing the efficacy of the independent variable.
Narrative Record: A form of data collection often used in Observational Research which tries to, more or less, provide a complete and faithful record of behaviors in the observed situation.
Naturalistic Observation: an attempt to observe behavior in as natural a state as possible.
Normal Distribution: Also called gaussian distribution. This distribution is symmetrical with the mean = the median = the mode. Approximately 68% falls between 1 standard deviation less than the mean and 1 standard deviation more than the mean.
Null Hypothesis: The statistical hypothesis that, in essences, there is not pattern or difference in the data. The particular form of the null hypothesis depends on the inferential statistic. In tests of differences, the null hypothesis is not difference between the groups. In tests of relationship, it is that the variables are not related.
Observational Design: A research design where a situation or event is observed and recorded in a systematic and objective fashion.
One-tailed test: A form of a statistical test where the direction of the outcome is predicted. See two-tailed test for comparison.
Order Effects: a confound in a within subject experiment where the order of the conditions, or stimuli influence subject responses.
p value: A value returned from a statistical test that indicates the probability that the null hypothesis is true, assuming that all the assumptions of the statistical test have been met.
Parsimony, Principle of: The idea that if two theorys both explain something equally well, you ought to use the simpler of the two theories.
Participant Observation: An observational method where the observer actually lives with the group being observed.
Pearson's Product Moment Correlation Coefficient: A linear correlation coefficient that measures the strenth of relationship between two or more variables use the least-squares method.
Pilot Study: A preliminary study performed to validate methods and measures.
Plagerism: Taking credit for work done by other people.
Population: All members or values of interest. The complete set to which the sample is thought to generalize.
Power: In a statistical sense, it is the probability of rejecting a false null hypothesis.
Predictive Validity: The measure of validity where the psychological measure is correlated with some behavior that occurs at a later point in time.
Probability Sample: probably a better term for a random sample.
Proposition: A claim about human behior or nature that forms a part of a theory.
Random Assignment: Where each participant has an equal chance of being in any of the different conditions of an experiment.
Random Sample: A sample where every member of the population has an equal chance of being selected.
Reactivity: In observational methods the idea that the behavior of the participants may be affected by the act of being observed.
Reductionism: The theoretical position that all phenomena are explainable in terms of the most basic forces of nature. Opposed to emergent properties.
Reliability: When refering to a measurement technique, it is the degree to which the measurements can be repreated and the same values obtained. See also: Inter-rater reliability, test-retest reliability, and split-half reliability.
Replication: The idea that scientific findings need to be repeated to be accepted.
Research Hypothesis: This is the hypothesis that is being tested in a particular piece of research. To be a scientific hypothesis is must meet the principle of falsification.
Response Set: The response on one trial of a study is influence by responses made on preceding trials.
Sample: A subset of all the values of interest.
Sample, Random: see random sample.
Sampling Distribution: The probability distribution of a sample parameter (such as the mean) would take if repeated samples of the same size were taken from the same population.
Significant: In a statistical sense, the proability of the null hypothesis being true, the p value, is lower than the cutoff probability, the alpha level.
Single-Blind Study: A study where the participants do not know the nature of the treatment or conditions, but the researcher does.
Situation Sampling: Recording behavior in different situations.
Split-half reliability: A technique to determine the reliablity of a measurment instrument by correlating half of the scrores against the other half. Usually done with Pearson's correlation coefficient.
Standard Deviation: A measure of variation of a set of values. It is calculated as the square-root of the mean squared deviations. If the values are normally distributed, the standard deviation and the mean can completely describe the set of numbers.
Standard Error: The standard deviation of the sampling distribution of the mean. Usually calculated as Standard Deviation/Square root (N).
Statistical Hypothesis: A hypothesis about data used in an inferential statistic. This hypothesis is about patterns, relationships, or differences in the data themselves. See Null Hypothesis and Alternative Hypothesis.
Statistical Power: see Power.
Structure Observation: Observational method where a stituation is set up by experimenter and then the behavior of the participants is observed without further interference. Used to study rare situations.
Survey: A paper and pencil instrument that examines any issue or phenomenon, e.g. voter surveys.
Syllogism: A syllogism is a deductive argument where a conclusion is reached from two premisses.
Test: A survey that measures or attempts to measure a psychological construct.
Test-retest reliability: A measure of the degree to which two administrations of the same measurement technique yeild the same scores for the same participants. Measured with Pearson's correlation coefficient, usually.
Theory: A logically organized set of claims about the issue under investigation which serve to define the events and relevant variables, describe the relationship among variables, and eplain the events and relationships.
Time Sampling: In Observational Research, recording the behavior at different times. These can be randomly or systematically selected.
t-test: This is a statistical test used to compare means when the sampling distribution of the means is one of the family of t-distributions and not the normal distribution. Some of the types are:
One-sample t-test: A mean sample is compared against a known population mean when the standard deviation of the population is not known. Compare to z test.
Two-tailed test: A type of statistical test where the direction of the outcome is not predicted. See one-tailed test for comparison.
Type I error: Rejecting a true null hypothesis.
Type II error: Retaining (not rejecting) a false null hypothesis.
Validity: That a measurment technique measures what it is intended to measure. See also: Convergent validity, discriminant validity, criterion validity, concurrent validity, and predictive validity.
Variable: Is anything that can assume more than one value. For example, age or gender.
z statistic: This value indicates the number of standard deviations a value is from the mean.
z test: An inferential
statistic to test if a sample comes from a known population. To use the z test,
both the population mean and standard
deviation must be known.
since August 13, 1998