Research and Statistics.txt

An organized process for collecting knowledge

A group of logically related statements that explains things that have occurred in the past and predicts things that will occur in the future.

A set of steps followed by scientists to ensure a common basis for conducting research.

Research in which no manipulation of variables is involved an no cause-and-effect relationship is studied.

Research that describes the characteristics of a phenomenon without attempting to determine what causes the phenomenon.

A method of research used to determine relationships between two or more variables.

Research that examines phenomena within the cultural and social context in which it takes place.

Research in which a cause and effect is unambiguously tested.

Research that is done when groups are preassigned to "treatments," such as gender, social class, and neighborhood, also known as Post Hoc Research.

Research that examines cause-and-effect relationships through the use of control and treatment groups.

The method and structure of an investigation chosen by the researcher to conduct data collection and analysis.

Pure research which adds to the base of information in a field but has no immediate application.

Research that has an immediate application at the time it is completed.

Represents a class of outcomes that can take on more than one value.

Assignment of values to objects, events, or outcomes according to rules

The measure that reflects the outcomes of a research study

A variable controlled by the researcher in an attempt to test the effects on some outcome, the dependent variable.

A research design in which more than one independent variable is studied in various combinations with others.

When variables compete to explain the effects found in a study.

A variable that has a potential influence on the dependent variable.

A variable that has an unpredictable impact upon the dependent variable

A variable that is related to the variables of interest (such as the dependent and independent variable), masking the true relationship between the independent and dependent variable.

A statement of equality between groups in an investigation; the state of affairs that is accepted as true in the absence of other information. A starting point for observing the effects of the independent variable(s) on the dependent variable and as a benchmark for the comparison of chance versus significant differences between groups.

A statement of inequality between groups in an investigation. Suggestion of directional or nondirectional relationships between groups.

A research hypothesis that posts an inequality (such as a difference between groups) but makes no suggestion of the direction of that difference (such as more than or less than).

A research hypothesis that posits an inequality between groups with direction to that difference (such as more than or less than).

The entirety of some group

A representative portion of a population

The ability to draw inferences and conclusions from data

The degree of risk you are willing to take that you will reject a null hypothesis when it is actually true

The amount of risk one is willing to take that the null hypothesis is true even though it is rejected.

General information usually available through newspapers

Secondhand sources of historical data, such as newspaper clippings and summary statistics

People or documentation, which presents firsthand information.

A brief summary of journal article which appears before the actual article or in a collection of abstracts.

Consistency in performance or prediction

The most frequently occurring score

The most frequently occurring score

The score at which 50% of the scores in the distribution fall above it and 50% fall below it.

The extent to which the results of an experiment can be generalized.

The sum of all the scores in a distribution divided by the number of observations.

The accuracy in concluding that the outcome of an experiment is due to the independent variable.

The distance between the highest and lowest score in a distribution.

Average distance of each score in a distribution from the mean.

The truthfulness or accuracy within the score of a test or interpretation of an experiment.

The type of sampling used when the likelihood of selecting any one member of the population is known

When the likelihood of selecting any one member of the population is unknown.

A sampling procedure allowing for the equal and independent chance of subjects being selected as part of the sample.

An unbiased criterion used in the selection of subjects for a sample.

A random sampling procedure in which increments determine who becomes part of the sample; for example, every third person is selected.

The process of selecting a sample that represents different groups or levels of a population.

A random sampling procedure used when subjects are known to be unequal on some variable in the population.

A probability sampling procedure wherein units of subjects are selected, rather than the subjects themselves.

A nonprobability sampling procedure wherein the selected sample represents a captive audience; for example, sophomore college students in an introductory psychology class.

A nonprobability sampling procedure similar to stratified random sampling in that a particular stratum is the focus; however, a specified number is set to be selected, and once that number is met, no further selection occurs.

A stratified random sampling procedure wherein subjects in the sample are selected in proportion to how they are represented in the population.

The magnitude of the difference between the characteristics of the sample and the characteristics of the population from which it was selected.

procedures that allow inferences to be made from a sample to the population from which the sample was drawn

The scale representing a hierarchy of precision on which a certain type of variable might be assessed.

Measurement that assigns labels that do not suggest quantity

measurement that assigns only rank order to outcome.

measurement that assigns values representing equal distances between points but that does not allow for proportional comparisons.

measurement that allows for proportional comparison and a meaningful zero.

a variable that has an underlying continuum that can take on any value.

a variable that can take on one of several mutually exclusive values.

consistency in performance or prediction.

True score plus error score

The actual score for someone on some test.

The part of an individual’s observed score that is attributable to method or trait variance or error.

the part of an individual’s error score that is due to characteristics of the test or test taking situation

the part of an individual’s error score that is attributable to characteristics of the individual

an index of the strength of a relationship between two variables. It ranges in value from -1.00 to 1.00. it can be positive or negative

numerical index of the relationship between a set of variables

a measure of how stable a test is over time.

examines consistency between forms. The relationship of two tests made from the same pool of items.

a measure of the consistency from rater to rater. Consistency of results produced by the same test given by different people

examines how unified the items are in a test or assessment. A measure of reliability which examines the unidimensional nature of a test

indicates the extent to which a test represents the universe of items from which it is drawn. The extent to which a test fairly represents the universe of all possible questions that might be asked

how well a test estimates (concurrent validity) or predicts (predictive validity) performance outside of the testing situation; a measure of the extent to which a test is related to some criterion

the extent to which the results of a test are related to an underlying psychological construct. The extent to which a test truly measures a proposed psychological ability of skill and is related to an underlying theory or model of behavior

various traits are measured using various methods. Regardless of how they are measured the scores are related. Thus, if the same trait is measured using different methods, the scores should be related, and if different traits are measured using the same methods, the scores should not be related.

a component of construct validity in which method variance is shared when measuring the same traits.

a component of construct validity in which trait variance is shared when using the same method.

test used to measure knowledge of a specific area

tests with standard instructions and scoring procedures which are used for all administrations of the test. One type of achievement test

Tests designed for a specific purpose with specific scoring and instructions for that purpose. One type of achievement test.

tests allowing you to compare individual’s test performance to the test performance of other individuals. A test in which the individual’s performance is compared with the results of a larger group of peers.

tests defining a specific criterion or level of performance, and the only thing of importance is the individual’s performance, regardless of where that performance might stand in comparison with others. A test measures mastery of specific definitions of performance for an individual in a particular content domain.

a process of evaluating multiple choice items by using difficulty level and the ability of the item to discriminate or differentiate between group performance

the proportion of test takers who got the multiple choice item correct.

an index that describes how well a multiple choice item differentiates between high scores and low scores on a test

tests which assess an individual’s feelings about an object, person, or event

a method of measuring attitudes. A method used in constructing attitude tests in which all of the items are assigned an attitude score. It is made up of nearly equal intervals for individuals to agree or disagree with various statements

Thurston Scale

a method used in attitude scales that requires the individual to agree or disagree to a set of statements using a five-point scale

Likert Scale

tests assessing stable individual behavior patterns

personality tests that ask the participant to respond to an ambiguous stimulus. It is assumed that the participant will project their worldview onto the stimulus.

interview questions that have a clear and apparent focus and a clearly called for answer (close-ended questions)

the researcher uses a device to keep track of time and measures the length of time that a behavior occurs. Recording behavior based on the amount of time it lasts.

Recording behavior based on the incidence or frequency of the occurrence of a particular behavior

a particular subject is observed/recorded during a particular interval of time

all of the behavior of the target subject is recorded, with little concern as to the specificity of its content. Recording behavior on a continuous basis.

Each score for each individual on a test or in an experiment.

data that are unorganized

A form used to record raw data and often used to facilitate entry in to the computer.

A specially printed scoring sheet that can be read and scored by computer.

A special computer that reads optical scoring sheets

using numbers to represent data

Simple measures of a distribution’s central tendency and variablility.

Set of tools to help you make decisions about how the data you collected relate to your original hypotheses and how they might be generalizable to a larger number of subjects than those who were tested. Procedures that allow inferences to be made from a sample to the population from which the sample was drawn.

The general shape of data which includes a mean, median, and mode

measures of central tendency represented as the mean, median, or mode.

scores that have the same reference point and the same standard deviation. Scores that have been derived to create a common reference point and the same standard deviation to allow for easy comparison.

It is the occurrence of variability which cannot be accounted for by any of the variables that you are studying. The unassuming explanation for differences between groups that implies that the differences are accounted for by variables other than those being studied.

dictates that regardless of the shape of the distribution (be it normal or not), the means of all the samples selected from the population will be normally distributed. The theorem in inferential statistics which states that regardless of the shape of the population distribution, repeated samples from it will produce means that are normally distributed

Same as statistical significance

Inadvertent acceptance of a false null hypothesis. The acceptance of a false null hypothesis. The probability that a Type II error will occur can be reduced by increasing the size of the sample.

The application of a statistical procedure to determine whether observed differences exceed the critical value, indicating that chance is not the most attractive explanation for the results.

computation of the test statistic value. The value obtained by applying a statistical test of significance.

the minimum value you would expect the test statistic to yield if the null hypothesis is indeed false. The tabled value at which point the null hypothesis cannot be accepted; the minimum value you would expect the test statistic to yield if the null hypothesis is true.

The leeway for variation a statistical value has; they help determine the critical value of the test statistic; a reflection of the sample size.

an advanced technique that examines whether group differences occur for more than one dependent variable. Statistical procedures used to examine group differences that occur on more than one dependent variable.

an advance technique that allows the researcher to reduce the number of variables that represent a particular construct and then use factor scores as dependent variables. An advanced statistical technique that allows for the reduction of variables representing a particular construct and then uses factor scores as dependent variables.

a procedure that allows for the examination of trends and patterns that may exist in many different groups in many different studies.

The notion that the stronger the effects of a treatment, the smaller the required sample size.

“Sample surveys” examine the frequency and relationships between psychological and sociological variables and taps into constructs such as attitudes, beliefs, prejudices, preferences, and opinions.

Basic tool in survey research. A method of collecting data that is similar to an oral questionnaire. An interview can be informal and flexible or structured and focused

Neutral information about the respondent (age, living arrangements…).

Questions have a clear and apparent focus and call for an explicit answer. Interview questions that have a clear and apparent focus and a clearly called for answer.

Questions allow the interviewee to elaborate upon responses. Interview questions that provide a broad opportunity for the participant to respond.

a general plan for survey research of what activities will occur when.

When the interviewer subtly biases the respondent to respond in one way or another. Bias introduced when the interviewer subtly influences the interviewee’s responses.

describes the linear relationship between two or more variables without any hint of attributing the effect of one variable on another. A method of research used to determine relationships between two or more variables.

an index of the relationship between variables.

A plot of scores or data points which indicates the relationship between variables.

the squared correlation coefficient, which indicates the amount of variance in one variable that is accounted for by the other.

the portion of unexplained variance. The amount of variance that is unaccounted for in the relationship between variables

methods of research that examine changes over time

receives treatment

does not receive treatment

Quasi-Experimental Designs

not characterized by random selection of participants from a population, nor do they include a control group. Research designs that are characterized by a lack of random selection and assignment

A type of experimental design in which one group receives only one test

A type of experimental design in which one group receives both a pretest and posttest.

a true experimental design with a high degree of internal validity.

a true experimental design with a high degree of internal validity in which posttests are the only measures taken

a traditional experimental design in which there are four different groups of participants, and many different questions can be answered simultaneously with some relatively simple comparisons.

if what you see is a function of what you did…

the quality of an experimental design such that the results can be generalized from the original sample to another sample and then, by extension, to the population from which the sample originated.

a statistical tool that equalizes any initial differences that might exist.

A pre-experimental design in which groups are not equivalent at the beginning of the research and which generally lacks a suitable degree of internal validity. One of the most commonly used quasi-experimental designs.

A pre-experimental design with limited internal validity. It is similar to nonequivalent control group design except there is no pretest

Observing one subject over a variety of behaviors… allow for an in-depth examination of specific behaviors.

Level of behavior associated with a subject before an experiment begins

Methods of research that examine changes over time.

assess changes in behavior in one group of subjects at more than one point in time; A method of developmental research that assess changes in behavior in one group of subjects at more than one point in time.

examines several groups of people at one point in time; a method of developmental research used to examine age differences rather than age changes