Effect Size

Difference between two means in terms of standard deviations

Placebo Group

experience exactly the same conditions as the experimental group expect for the level of the independent variable is thought to affect the dependent variable.

NonParametric Test

significance test that does not make estimations of parameters of and underlying distribution.

Independent Variable

one group of participants is subject to the experimental condition while an entirely different group of people participates in the control condition. These two conditions are the levels of the independent variable that was manipulated by the experiment.

Deduction

Logical argument using rules to derive a conclusion from premises

Continious Scale

there is no limit to the sub divisions of points that are possible and meaningful, though there will be limits in practice, depending on the type of measuring instrument used

Data Set

group of data points or values which can be summarized or analyzed

Snowball Sample

a technique often employed in qualitative research where the information is required from key people in an organization or from those whose have personal or professional experience with the research issue.

Central Tendency

formal term for any measure of the typical or middle value in a group.

Pretest

Measure of participants before an experiment in order to balance or compare groups, or to asses changes by comparison with scores after the experiment.

Hypothesis testing

research that analyses data for a predicted effect

Order effects

effects from the order in which people participate in conditions.

Dispersion

technical and general term fro any measure of the spread of scores in a sample of the population data.

Related design

a score in one condition is discretely paired with a score in the other condition

Categorical Variable

each value id a discrete and qualitatively separate categories

Measured Variable

We can use a number to indicate where along a scale a person lies on this variable

Unrelated design

the 'independent samples' or 'between groups'. Each score in one group or condition cannot in anyway relate to any specific score in the other group or condition.

Range

the distance between the top and bottom values of a set

Reification

the tendency to treat abstract concept as if it had independent existence.

Factors

another name for independent variables

Construct Validity

the extent to which our measure actually coincides with the construct

Interquartile range

distance between the first and third quartile in the distribution

Operational definition

definition of phenomenon in terms of the precise procedures taken to measure it.

Hypothetical construct

phenomenon or construct assumed to exist and used to explain observed effects, but as yet unconfirmed; stays as an explanation of effects while evidence supports it.

Asymmetrical order effect

an order effect that has greater strength in one particular order and where, therefore, counterbalancing would be ineffective

Semi inter quartile range

half the distance between first and third quartile in a distribution

Reliability

The extent to which findings or measures can be repeated with similar results. Consistency of measures

Baseline Comparison

Measure of what would occur if no experimental level of the independent a

Deviation value/ score

the amount by which a particular value deviates from the mean

Validity

the extent to which instruments measure what they intended to measure. Also the extent to which a research effect can be trusted as real or as not 'contaminated' or confounded.

Population

all the existing members of a group

Sample

A group selected form the population for an investigation

Simple Random Sample

every case in the target population has an equal chance of selection and so does every possible combination of cases

Systematic Random Sample

select every nth case form the population where n is an number. before selection is started every case must have an equal chance of being in the sample

Sampling Frame

the population from which we will sample from

Factorial Design

experiment in which more than one independent variable is manipulated

Mean Deviation

measure of dispersion mean of all absolute deviations

Nonequivalent Groups

A possible confounding variable where two or more groups in a independent samples design experiment differ on a skill or characteristic relevant to the dependent variable

Standard Score

number of standard deviations a particular score us from it's sample mean.

Sampling Distribution

Theoretical distribution that would be obtained by taking the same statistic from many same size samples

Stratified Sample

A sample is selected so that specified groups will appear in numbers proportional to their size in the target population; within each subgroup cases are selected randomly.

Cluster Sampling

groups in the population selected at random from among other similar groups and assumed to be representative of a population.

Sampling Error

difference between a sample statistic and the true population statistic, usually to assumed to be random in origin

Degrees of freedom

common term in statistical analysis having to do with the number of individual data points that are free to vary given that overall summary values are known.

Standard error

the distribution has a standard deviation

Quota Sampling

Consists of obtaining people from categories, in proportion to their occurrence in the general population but the selection from each category is entirely determined by the researcher

Self selecting Sample

researcher use participants that are readily available

Variation ratio

proportion of nonmodal values to all values

Central Limit Theorem

Used in theoretical estimations of standard error of sampling distribution from standard deviation of a sample.

Opportunity/Convenience Sample

a sample selected because they are easily available for testing

Confidence Intervals

limits to the likely interval within which a population mean lies based on an estimate from a sample mean and standard error

Haphazard Sample

A sample selected from the population with no conscious bias ( but not likely to be truly random)

Sample statistics

measures of a sample. they are frequently taken as an estimate of the same measures of a population

Purposive Sample

The selection choice is made by the researchers on the basis of those who are more representative for the research issue or who are likely to have expertise in the matter

Line charts

chart joining continuous data points in a single line

Skewed Distribution

nonnormal distributions that have a lot more scores on one side of the mode than the other

Natural Experiments.

events beyond the researcher's direct control but where an Iv and DV can be identified.

Participant Variable

Variations between persons acting as participants & which are relevant to the study

timeseries

line chart showing measures of variable at progressive time intervals.

Sampling Bias

The weighting of a sample with an overrepresentation of one particular category of people

Ceiling Effect

occurs where a measure produces most values near the top end of a scale

Floor Effect

phenomenon where measures produce very many low scores

Group difference studies

a post facto study that compares the measurements of an existing variable in two contrasting conditions.

Kurtosis

the overall shape of a distribution in terms of height and width compared with normal distribution

Frequency Polygon

histogram showing only the peaks of class intervals

Platykurtic

nonnormal distribution that is widely spaced out and low in the centre

Leptokurtic

Nonnormal distribution that is closely bunched in the center and tall

Significance test

test performed in order to decide whether the null hypothesis should be retained or rejcted

Explanatory data analysis

close examination of data by a variety of means before submitting them to significance testing

Null Hypothesis

Assumption of no effect in the population from which the samples are drawn

Alternative Hypothesis

assumption that an effect exists

Mundane realism

feature of design where experiment resembles everyday but is not necessarily engaging

Participant Variable

person variables differing in proportion across different experimental groups and possibly confounding results

standardized procedure

Testing or measuring behavior with exactly the same formalized routine for all participants

Experimenter realism

effect of attentiongrabbing interesting experiment in compensating for artificiality or demand characteristics

Cumulative frequency

distribution that shows the number of cases that have occurred up to and including the current category

Significance Level

levels of probability at which it is agreed to reject the null hypothesis. if the probability if obtained results under the null hypothesis is less than the set level then the null hypothesis is rejected

Error Variance

Variance among scores caused by the operation of randomly acting variables.

Box plots

based on ordinal measurements of the set of data. The give us a graphical representation of the interquartile range

Experimental Validity

The truth of conclucions drawn about an experiment.

Deception

Leading participants to believe tat something other than the true independent variable is involved or withholding information such that the reality of the investigative situation is masked or distorted

Raw data

the unprocessed data gathered directly from participants before they are organized or treated statistically.

Frequency Distribution

distribution showing how often certain values occurred in categories

Debriefing

informing participants about the full nature and rationale of the study they've experienced and attempting to reverse any negative influences

Threats to Validity

any influence on research variables that might provide an alternative explanation of our effect or that might limit generality of what we appear to have found.

Stress and Discomfort

psychological investigators should guarantee the safety of their participants and that everything possible should be done to protect them from harm of discomfort

Percentile

point of on a measured scale that marks off a certain percentage of cases in a data set

Decile

Point on a measured scale that marks off each 10% of the data set of the population

The Right to Withdraw

in all research that involved individual participation the investigator is obliged to  1. give participants at to likely levels of discomfort, 2. remind them of their right to withdraw & 3. terminate the experiment if discomfort levels are higher than anticipated

Normal Distribution

continuous distribution, bell shaped, symmetrical about its midpoint.

Categorical weekend

variable where cases are merely placed into independent, separate categories

Informed Consent

agreement to participate in research in the full knowledge of the research context and participant rights

Construct Validity

whether our measures of a construct are really measuring it, and also the extent to which the construct itself is supported by psychological evidence

Measured Variable

variable were cases measured on it are placed on some sort of scale that has a direction

Participant expectancy

the expectation on the part of the participants that their behavior should alter in an experimental situation

Demand Characteristics

cues in a study which help participants work out what is expected

Social desirability

the tendency of research participants to want to look good and provide socially acceptable answers.

Anonymity

keeping participants or clients identity away from publication or any inadvertent disclosure

Evaluation apprehension

participants concern about being tested, which may affect results.

Reactive

studies in which the participant is expected to react to being studied

Confidentiality

keeping data from participants or clients away from publication

Single Blind

participants do not know what condition they are in

Nominal Level

level of measurement at which numbers are the only labels for the categories

Double blind

an experimental procedure whether neither the participants or the experimenter knows what condition the participant is in

Expectancy control design

half of the participants receive the experimental treatment and half do not as in the standard between groups design.

Intervention

research that makes alteration to peoples lives beyond the specific research setting in some cases because there is an intention to ameliorate specific human conditions

Independent Variable

The thing that is manipulated

Dependent Variables

The resulting behaviour that is measured

True Experiment

Participants are randomly assigned to conditions

Quasiexperiments

Participants are already on groups

Between Participants

divide the set of participants into two group. Each group has a different condtion

Within Participants

Each participant does the task twice. Once in each condition

Population Validity

the extent which an effect can be generalized from the sample to the sample population from which they were selected and to other populations

Frequencies

number of cases in specific categories. These numbers are being used to count, they do stand for quantities.

Onetailed Hypothesis

X will be bigger/smaller than Y

Critical Value

value that the results of our test statistic must reach in order for the null hypothesis to be rejected.

Two tailed Hypothesis

X will be different to Y

Type II Error

The probability of retaining the null hypothesis when it is false

ordinal level

level of measurement at which cases are arranged in rank position

Ecological Validity

the extent to which findings generalize to other settings

Internal validity

when we can conclude that the independent variable brought about the observed changes in the dependent variable

Effect Size

the size of the effect being investigated as it exist in the population

Interval level

level of measurement at which units on a scale represents an equal change in the variable measured

External Validity

The degree to which we can generalise our results to other participants, other conditions, other times and other places

Confounding Variables

A variable that systematically changes with the independent variable and dependent variable

Power

The probability of not making a type II error if a real effect size occurs. The probability of selecting a case or sample above the level of Beta in the population defined by the alternative hypothesis id 1  Beta

Metaanalysis

Statistical analysis of results results of multiple equivalent studies of the same or very similar hypotheses in order to assess more thoroughly an effects validity

One tailed Test

test referring to only one tail of the distribution under the null hypothesis. May be used i alternative hypothesis is directional

Hawthorne effect

effect on human performance caused solely by the knowledge that one is being observed

Two Tailed hypothesis

test referring to both tails of the probability distribution. Must be used if the alternative hypothesis is nondirectional

Critical Case

A special case may sometimes highlight things that can be related back to most nonspecial cases

Lab Experiment

conducted in a highly controlled artificial environment. There is a problem of low external validity

Parametric test

Relatively powerful significance test that makes estimations of population parameters. The data tested must usually therefore satisfy certain assumptions.

Field experiment

conducted in naturally occurring environment for participants

Focus Groups & Panels

Researchers may bring together a panel of either experts, those with a particular interest in a topic or in the case of focus groups, simply a selection of people who are fairly representative of the general population.

Distribution Dependent Test

significance test making estimations of population parameters.

Placebo

Inert innocuous version of treatment resulting in psychosomatic placebo effect.

Quasiinterval scales

scale that appears to be interval but where equal intervals do not necessarily measure equal amounts of the construct.

Sampling Bias

Systematic tendency towards over or under representation of some categories in a sample

Difference Mean

mean of differences between pairs of scores.

Ratio

interval type scale where proportions on the scale are meaningful because there is an absolute zero point

Pooled Variance

combining of two sample variances is to average in order to estimate population variance

Positivism

Methodological belief that the world's phenomena are reducible to observable facts and the mathematical relationships between them.

Control Groups

groups used as a baseline measure against which the performance of the experimental group is assessed

Discrete Scale

each point is entirely separate from the next
