Definition of Terms



abstract population: The population about which one ultimately wishes to draw conclusions.

action research: Applied research using the scientific method.

analysis of variance (ANOVA): A procedure for testing hypotheses about relationships among variables under a wide range of conditions for three or more groups.

applied research: Research emphasizing the solution of social and community problems.

attitude scale: A type of questionnaire designed to produce scores indicating the overall degree of favorability of a person's attitude on a topic, such as free time.

attribute: One of the categories specified or implied by a variable.


basic research: Research emphasizing the solution of theoretical problems.

binomial probability distribution: The probabilities associated with every possible outcome of an experiment involving n independent trials and a success or failure on each trial.

bivariate analysis: The analysis of relationships among pairs of variables.


census: A complete count of the members of a population; also refers to the count and survey process conducted by the U.S. Census Bureau.

central tendency: The degree to which the quantities of a variable converge.

cluster analysis: An exploratory procedure for combining similar objects into groupings and then using those groupings ("clusters") as a basis for further analysis.

cluster sample: A probability sample in which groupings of elements are selected initially and individual elements are sampled subsequently.

code: A translation from raw data to numbers that can be employed by data entry personnel.

coding: A procedure that assigns data to numbered categories in order to facilitate quantitative and qualitative analysis.

coefficient of determination: A measure of the proportion of the variation in the dependent variable that the independent variable is able to account for.

complete observer: An observational role in which the researcher's behavior is not part of the phenomenon studied and subjects remain unaware of the study and observer.

complete participant: An observational role involving as thorough as possible an entrance into the lives of those being studied.

concept: A linguistic symbol that categorizes phenomena.

concurrent validity: The correlation of results correspond with other results.

construct validity: The degree to which an operational definition categorizes phenomena in the same way as a well-accepted or criterion operational definition. Simply, do items measure hypothetical constructs or concepts?

content analysis: Research procedures for relating symbolic data to their context; systematically describing the form and content of written or spoken material.

content validity: The condition in which the items on the survey measure the content they were intended to measure.

control group: A group not subjected to the experimental treatment, which is used for purposes of comparison. The control group resembles the experimental group in every respect except that it is not exposed to the independent variable(s) in order to control for the effects of extraneous variables on the dependent variable.

controlling on a variable: Crosstabulating other variables within each of the attributes of the variable, such as female and male, age, etc..

convenience sample: A nonprobability sample drawn primarily on the basis of the ease of obtaining the data.

correlation: A measure of the degree of fit (a coefficient indicating the relationship) between two sets of scores (variables).

corroboration: Confirmation of statements from two or more sources.

cross-tabulation: The distribution of one variable within the categories of another.

cumulative scaling procedures: Techniques for constructing a partially ordered scale that include some testing of that scale's assumptions.


data: Information about the nature of phenomena derived from experience or observation.

degree of relationship between two variables: The closeness of the relationship between the variables, usually measured by a correlation coefficient.

demographics: a variety of socio-demographic characteristics that can be identified and used to categorize groups with shared behaviors or traits including: age; gender; ethnicity; income; and education.

dependent variable: A variable presumed to be influenced by another variable.

discriminant analysis: An exploratory procedure for identifying relationships between nominal-scale dependent variables and more quantitative independent variables, often used to describe group characteristics.


elaboration model: An image of bivariate and multivariate cross-tabulation procedures for making sequential inferences from one-shot data.

ethnomethodology: A theoretical orientation emphasizing people's everyday procedures for solving practical problems in situations through 

(1) constructing ethnotheories, 
(2) experiencing phenomena, and 
(3) constructing common-sense accounts and prophesies.

evaluation research: Applied research using the scientific method to assess the worth or effectiveness of a leisure activity, program, or policy.

ex post facto hypothesizing: Developing hypotheses to fit observed data.

experiment: A method of data collection in which the researcher tests hypotheses by introducing a change in the research situation and observing the results.

experimental group: A group subjected to the experimental treatment.

experimental post test: Observation subsequent to introduction of the experimental treatment.

experimental pretest: Observation prior to introduction of the experimental treatment.

experimental treatment: The change that the experimenter introduces.

exploratory observation: A type of pretest centering on the method of noting and recording ongoing phenomena.
external validity: The degree to which statements about the specific phenomena investigated (the sample) can be generalized to other settings population).


face validity: The degree to which an operational definition, on the basis of the researcher's experience and that of other experts, appears to correctly specify the concept it is designed to specify. Specifically, do the items appear to measure what the instrument purports to measure?

factor analysis: An exploratory correction procedure for deriving a relationship among a small number of "factors" or dimensions from a larger )f variables; useful for data reduction, explanation of dimensions, and scaling.

field experiment: An experiment within an ongoing setting.

field observation: The noting and recording of behavior within an ongoing setting.

fixed-alternative question: A question that provides a list of alternative answers with a restrictive scale format (yes/no; strongly agree/strongly disagree).

focused interview: An interview centering on the effects of a given phenomenon experienced by the respondent.

funnel technique: The use of successively more structured and probing questions.


goodness of fit: Degree to which observed data coincide with theoretical expectations.

grand mean: the means of all the scores in an experiment.

grouped frequency distribution: An arrangement of  scores from highest to lowest in which scores are grouped together into equal-sized ranges called class intervals.


historical method: The process of critically examining the records of the past.

historiography: The imaginative reconstruction of the past by means of historical method.

hypothesis: A tentative statement of a relationship between two or more variables.


independent variable: A variable presumed to influence or precede another variable (dependent variable). The variable is systematically manipulated by the researcher to determine changes in the dependent variable. Also known as the experimental variable or moderating variable.

internal validity: The degree to which the instrument (statements) or procedure in a study measures what it is supposed to measure.

interval scale: A measurement of a variable which results in the classification of phenomena into a set of attributes with equal distances or intervals separating them (l=Strongly Agree, 2=Agree, 3=Neutral, 4=Agree, 5 =Strongly Agree).

intervening variable: A variable that follows the independent variable and precedes the dependent variable, sometimes manipulated to determine effects of the independent variable on the dependent variable through the means of covariation.

investigator effect: The impact of the scientist on the research process.


j-curve: A severely skewed distribution with the mode at one extreme.


laboratory experiment: An experiment within a deliberately constructed setting.

laboratory observation: The noting and recording of behavior within a deliberately constructed setting.
Likert-type scales: Measurements employing many of the criteria specified by Likert for summated scaling procedures, usually a three-, five-, or seven-point scale.


measurement: The process of creating a correspondence between a concept and data in specifying that concept.

method of agreement: An experimental method that designates a presumed partial cause of a phenomenon as any factor that occurs together with the phenomenon.

method of concomitant variations: An experimental method that designates as a presumed partial cause of a phenomenon any factor that varies in the same way that the phenomenon varies.

method of difference: An experimental method that designates a presumed partial cause of a phenomenon as any factor that occurs together with the phenomenon and does not occur when the phenomenon does not occur.

method of equal-appearing intervals: A technique for constructing an interval scale based on judges' ratings of a large number of items on the items' degree of "favorableness" to a given variable.

method of paired comparisons: A technique for constructing an interval scale based on judges' choices-for all possible combinations of pairs of items-of items on the basis of their degree of "favorableness" toward a given variable.

multiple correlation: A measure of the degree of fit between actual scores for a dependent variable and the scores predicted by a set of independent variables.

multiple regression: A prediction of the scores of a given dependent vari- able using the scores of a set of independent variables. Of interest is the relationship between the sets of scores. Can we predict attitude or behaviors based on the influence or presence of a set of variables or conditions (independent measures)?

multivariate analysis: The analysis of relationships among three or more variables at the same time. For example, in a leisure services marketing study, the effects of age, income, and education could be analyzed simultaneously to determine which of the variables would better predict (or explain the largest amount of variation in) nonuse of a program.


nominal scale: A measurement of a variable which results in the classification of phenomena into a set of consistent and non-overlapping attributes (yes, no, male, female, etc.).

nonparametric statistics: Tests of significance that require few assumptions about the population. Use of these statistics should occur when samples are small (fewer than 30 subjects), when subjects were not randomly sampled, or data are not interval level. Chi-square, Spearman rank order coefficient, and Mann-Whitney U are some examples.

non probability sampling: A method of selecting a sample from a population that does not yield known probabilities for selecting each sample element. There are three types: quota, purposive, and convenience samples.

null hypothesis: A hypothesis about a population made for the purpose of testing it against sample data; a conservative means of stating the research questions to indicate that the independent variable has no effect on the dependent variable.


objectivity: An unprejudiced or open orientation to information about the nature of phenomena being studied.

observation: The act of noting and recording a phenomenon, often with instruments (cameras, recorders) or note taking.

observer-as-participant: An observational role revealed to others and offering a very limited time for data collection.

open-ended question: A question that does not provide a list of alternative answers, but allows the respondent to express his or her position on a topic.

operational definition: An explicit procedure for defining a variable by the means used to measure it. The personal meaning of leisure can be operationally defined as the responses to a set of involvement profile questions.

ordinal scale: A measurement of a variable that results in the classification of phenomena into a set of ranked or ordered attributes. The ordinal scale has the characteristic of the nominal scale  (different numbers mean different things) plus the characteristic of indicating "greater than" or "less than." 


parametric statistics: Researchers use these statistics when samples are larger than 30 subjects and are randomly drawn, groups have equal variances, and data are interval. Examples of these statistics are the T-test and analysis of variance.

partial correlation: A procedure for measuring the degree of association between an independent variable and a dependent variable while con- trolling the influence of one or more other variables.

participant observation: A method of data collection in which the re- searcher notes and records ongoing social phenomena with the researcher's Own behavior constituting part of the phenomena.

participant-as-observer: An observational role involving the relation of that role to others as well as thorough entrance into their lives.

path analysis: A systematic procedure for constructing a casual model of the relationships among a set of variables, assuming linear relationships exist.

Pearson product-moment correlation coefficient (r): a measure of association based on least-squares deviations from a regression line. A widely used statistic for measuring the relationship between two sets of scores that are assumed to be continuously distributed.

population: A set of elements from which a subset may be drawn (a sample). It is the entire group of people, objects, or events in a category.

predictive validity: The condition in which scores predict a criterion measure.

pretest: A trial run of a data collection procedure to determine problem areas or weaknesses in design.

primary source: The testimony of an eyewitness, that is, someone present at the events described.

probability level: An indication of the likelihood that an obtained difference on a statistical test is due to chance alone. A common practice in leisure research, and behavioral and social science research, is to use the .05 probability level (also called the .05 level of significance).

probability sampling: A method of selecting a subset of elements (sample) from a larger set (population) in such a way that each element of the larger set has a known probability of being selected. Two general types are random and stratified samples.

purposive sample: A nonprobability sample chosen when individuals considered most closely related to the issue being studied are selected for inclusion.


quota sample: A nonprobability sample that takes into account the proportion of individuals in different population categories (age, gender, etc.) within the population. This is a nonprobability type of sample method.


randomization: The use of probability sampling to assign subjects to experimental and control groups.

ratio scale: A measurement of a variable that results in the classification of phenomena into a set of attributes. This scale is characterized by an absolute zero point as well as equal distance between attributes.

research triangulation: The use of multiple methods of investigation. The assumption for use of this method is that any bias inherent in the study would be neutralized in conjunction with other data sources and collection procedures. There are two types: 


within methods; and 


between methods. 

The within-methods approach utilizes only quantitative (surveys and experiments) or only qualitative (observation and interviews) approaches. The between-methods approach mixes both quantitative and qualitative methods.

regression: A prediction of the scores (or values) of a given dependent variable, based on the scores of one or more independent variables.

reliability: The degree to which an operational definition is stable and consistent under similar circumstances. Reliability amounts to consistency in measurement, the repeatability or replicability of findings. Specifically, reliability applies to measures of item consistency (are item responses consistent across constructs?), test stability (do responses vary when test is given a second time?); and consistency in administration and scoring (were errors caused by carelessness in administration or scoring?).

replication: The repeating of a study to determine if findings can be duplicated. This is important to increase confidence in research results and to establish models.

research ethics: Principles that guide an investigator's choices and procedures. Of concern is maintaining confidentiality of data, preserving anonymity of informants, and using research for the intended (stated) purposes.


sample: A subset of elements selected from a larger set (the population). The sample is used to make generalizations about the population from which it was drawn.

sampling bias: The error introduced by a sampling procedure that favors certain characteristics over others.

sampling error: The degree of inaccuracy of statements about a population due solely to differences between that population and a sample drawn from it; chance variation among samples selected from the same population.

scale: A series of ordered steps at fixed intervals used as a standard of measurement.

scaling: The process of constructing an operational definition with numerical properties. Examples in leisure include scales to determine attitudes toward free time, involvement, barriers, benefits, and tourism impacts. Factor analysis and reliability testing are used for the scaling process.

scattergram: A set of plots for data on two coordinates.

scientific method: A problem-solving process that involves 
(1) proposing theory, 
(2) using data to verify theory, and 
(3) creating more explanations and accurate predictions of the phenomenon.

scientific paradigm: The world view on which a science is based, involving a system of explicit and implicit assumptions.

secondary analysis: The analysis of available data within a framework that differs from that used in the original study.

secondary source: The testimony of anyone who is not an eyewitness.

semantic differential: A questionnaire procedure for measuring the meaning an individual attaches to a phenomenon using opposing adjectives (good to bad; bored to engaged). The scale was developed by Charles Osgood for measuring the meaning of concepts.

semistandardized interview: An interview involving a moderate degree of planned structure established prior to the interview.

simple random sampling: A probability sampling procedure in which each element of the population has an equal probability of being chosen.

spurious relationship: The condition where a third variable precedes, and is seen as a cause of, the independent and the dependent variables.

standard deviation: A measure of variation based on squared deviations from the mean.

standardized interview: An interview involving a high degree of planned structure.

statistical inference: A statement about a population based on procedures for analyzing a probability sample of that population.

stratified probability sample: A probability sample based on selections from different population categories (gender, age, occupation, voter/non- voter, etc.).

summated scaling procedures: Techniques for developing partially ordered scales by summing scores over a number of items.

survey: A procedure for systematically collecting information from people by obtaining their responses to questions.

systematic sample: A probability sample drawn from a list where every nth element is selected, usually after a random starting place in the list.


theory: Explicit ideas about the nature of phenomena. Theory is an inter-related set of constructs (or variables) formed into propositions (or hypotheses) that specify the relationship among the variables.

treatment: different levels of the independent variable.

type I error: The rejection of a correct null hypothesis.

type II error: The acceptance of an incorrect null hypothesis.


univariate analysis: The analysis of single variables as distinct from relationships among variables.

unstandardized interview: An interview involving little or no planned structure. Although the interviewer has a general topic in mind, there is no predetermined order or specified wording to the questions.


validity of measurement: An indication that the measure accurately reflects what it is supposed to measure (see also internal and external validity).

value communication: The researcher's communication of his or her own values.

value neutrality: The avoidance of all pronouncements of an ethical or value-laden character.

variability: The degree to which the quantities of a variable diverge. Variability is the amount of spread or dispersion within a distribution of scores. Common measures of variability are the range and the standard deviation.

variable: A concept specifying or implying more than one category to which phenomena may be assigned. Any characteristic or quality that differs in degree or kind. Variables may be continuous (1,2,...) or categorical (yes/no, etc.).

variance: the square of the standard deviation.


z-score: A score expressed in standard-deviation units: used to compare the relative standing of scores in two different distributions.


Copyright 2001. Northern Arizona University, ALL RIGHTS RESERVED