Printer friendly version

Accessible Population – The population of subjects available for a particular study; often a nonrandom subset of the target population.

Accidental Sample – Selection of the most readily available persons as subjects in a study; also known as a convenience sample.

Anecdotal Reports – Relating the experiences and clinical impressions of the author; of limited utility in literature reviews for research studies because of their highly subjective nature.

Attribute Variables – Preexisting characteristics of the item/subject under investigation that the researcher simply observes and measures.

Attrition – The loss of participants during the course of a study; attrition can introduce an unknown amount of bias by changing the composition of the sample.

Bias – Any influence that produces a distortion in the results of a study.

Blind – A study design in which the subjects are unaware of whether they are the treated or the control subjects.

Case-Control Study – A study design that is similar to cross-sectional studies in that it assesses the relationship of an attribute to another variable or attribute; cases (e.g., patients with a certain disease) are compared with a control or comparison group (e.g., those without the disease).

Case Study – A research method that involves a thorough, in-depth analysis of an individual, group, or institution.

Causal Relationship – A relationship between two variables such that the presence or absence of one variable (the cause) determines the presence or absence, or value, of the other (the effect).

Cluster Sampling – A form of multistage sampling in which large groupings (clusters) are selected first (e.g., AIDS patients), with successive subsampling of smaller units (e.g., AIDS patients with toxoplasmosis).

Cohort Study – A kind of trend study that focuses on a specific subpopulation (often an age-related subgroup) from which different samples are selected at different points in time (e.g., males aged 30 to 40 years).

Comparison Group – A group of subjects whose scores on a dependent variable are used as a basis for evaluating the scores of the experimental group; the term “comparison group” is generally used instead of “control group” when the investigation does not use a true experimental design.

Control – The process of holding constant the possible influences on the dependent variable under investigation.

Control Group – Subjects in an experiment who do not receive the experimental treatment and whose performance provides a baseline against which the effects of the treatment can be measured (see Comparison Group).

Convenience Sample – Selection of the most readily available persons as subjects in a study; also known as an accidental sample.

Correlation – A tendency for a variation in one variable to be related to a variation in another variable.

Correlational Research – Investigations that explore the interrelationships among variables of interest without any active intervention on the part of the researcher.

Covariate – A variable that is statistically controlled during analysis; the covariate is typically an extraneous, confounding influence on the dependent variable.

Criterion Variable – The quality or attribute used to measure the effect of an independent variable; sometimes referred to as the dependent variable.

Crossover – A study design in which the subjects can choose to leave the group to which they have been randomly assigned (i.e., experimental group or control group) and “crossover” into the other group; this leads to difficulty in discerning differences between the variables and in the assessment of the efficacy of an intervention.

Cross-Sectional Study – A study based on observations of different age or developmental groups at a single point in time for the purpose of inferring trends or changes over time; also known as prevalence studies.

Dependent Variable – The outcome variable of interest; the variable that is hypothesized to depend on or be caused by another variable (the independent variable); sometimes referred to as the criterion variable.

Dichotomous Variable – A variable having only two values or categories.

Direct Relationship – A positive correlation between two variables.

Double-Blind Experiment – An experiment in which neither the subjects nor those who administer the treatment know who is in the experimental or control group.

Effect Size – A measure of the strength of the relationship among variables. If the researcher has reason to believe that the independent and dependent variables will be strongly interrelated, then a relatively small sample is generally adequate to demonstrate the relationship statistically; when the researcher has no reason to believe that relationships will be strong, then relatively large samples are needed.

Event Sampling – In observational studies, a sampling plan that involves the selection of integral behaviors or events.

Experiment – A research study in which the investigator controls (manipulates) the independent variable and randomly assigns subjects to different conditions.

Experimental Group – The subjects in an experimental or quasi-experimental study who receive the experimental treatment or intervention.

External Validity – The degree to which the results of a study can be generalized to settings or samples other than the ones studied.

Extraneous Variable – A variable that confounds the relationship between the independent or dependent variables and needs to be controlled either in the research design or through statistical procedures.

Factorial Design – An experimental design in which two or more independent variables are simultaneously manipulated. This design permits an analysis of the main effects of the independent variables separately, plus the interaction effects of these variables.

Follow-Up Study – A study undertaken to determine the subsequent development of individuals with a specified condition or who have received a specified treatment.

Functional Relationship – A relationship or association between two variables wherein it cannot be assumed that one variable caused the other; however, it can be said that variable X changes values as a function of changes in variable Y.

Generalizability – The degree to which the research procedures justify the inference that the findings represent something beyond the specific observations on which they are based; in particular, the inference that the findings can be generalized from the sample to the entire target population.

Hawthorne Effect – The effect on the dependent variable caused by subjects’ awareness that they are “special” participants under study.

Heterogeneity – The degree to which objects/people are dissimilar with respect to some attribute.

Homogeneity – The degree to which objects/people are similar with respect to some attribute.

Hypothesis – A statement of a predicted relationship between the variables under investigation.

Independent Variable – The variable that is believed to cause or influence the dependent variable. In experimental research, the independent variable is the variable that is manipulated.

Informed Consent – An ethical principle that requires researchers to obtain the voluntary participation of subjects, after informing them of possible risks and benefits.

Institutional Review Board (IRB) – A group of individuals who convene to review proposed and ongoing studies with respect to ethical considerations.

Interaction Effect – The effect on a dependent variable of two or more independent variables acting in combination (interactively) rather than as unconnected factors.

Internal Validity – The degree to which it can be inferred that the experimental treatment or independent variable, rather than uncontrolled extraneous factors, is responsible for observed effects on the dependent variable.

Intervention – In experimental or quasi-experimental research, the experimental treatment or manipulation.

Inverse Relationship – A negative correlation between two variables.

Judgmental Sampling – A type of nonprobability sample method in which the researcher selects subjects for the study on the basis of personal judgment about which one will be most representative or productive; also referred to as purposive sampling.

Level of Significance – The risk of making a Type I error, established by the researcher before the statistical analysis.

Longitudinal Study – A study designed to collect data at more than one point in time, in contrast to a cross-sectional study.

Manipulation – An intervention or treatment introduced by the researcher in an experimental or quasi-experimental study. The researcher manipulates the independent variable to assess its impact on the dependent variable.

Matching – The pairing of subjects in one group with those in another group based on their similarity on one or more dimension, done in order to enhance the overall comparability of groups. When matching is performed in the context of an experiment, the procedure results in a randomized block design.

Mean – A descriptive statistic that is a measure of central tendency, computed by summing all scores and dividing by the number of subjects.

Median – A descriptive statistic that is a measure of central tendency, representing the exact middle score or value in a distribution of scores; the median is the value above and below which 50% of the scores lie.

Mode – A descriptive statistic that is a measure of central tendency; the score or value that occurs most frequently in a distribution of scores.

Multistage Sampling – A sampling strategy that proceeds through a set of stages from larger to smaller sampling units.

Negative Predictive Value – The probability that an individual with a negative test result does not have or will not develop the condition for which the test was conducted.

Negative Relationship – A relationship between two variables in which there is a tendency for higher values on one variable to be associated with lower values on the other; also referred to as an inverse relationship.

Nonequivalent Control Group – A comparison group that was not developed on the basis of random assignment. When randomization is not used, there is no way of ensuring the initial equivalence among different groups.

Nonexperimental Research – Studies in which the researcher collects data without introducing any new treatment or intervention.

Nonprobability Sampling – The selection of subjects or sampling units from a population using nonrandom procedures; examples include accidental, judgmental, and quota sampling.

Null Hypothesis – A proposition that a potential treatment has no effect and that any observed difference is due to chance.

Observational Research – Studies in which the data are collected by means of observing and recording behaviors or activities of interest.

Outcome Measure – A term sometimes used to refer to the dependent variable.

Panel Study – A type of longitudinal study in which the same subjects are used to provide data at two or more points in time.

Pilot Study – A small-scale version, or trial run, done in preparation for a major study.

Population – The entire set of individuals (or objects) having some common characteristic(s); sometimes referred to as universe.

Positive Predictive Value – The probability that an individual with a positive test result has or will develop the condition for which the test was conducted.

Positive Relationship – A relationship between two variables in which there is a tendency for high values on one variable to be associated with high values on the other.

Power Analysis – A procedure for estimating either the likelihood of committing a Type II error or sample size requirements.

Predictive Value – The likelihood that a diagnostic test measures the condition for which it was conducted.

Preexperimental Design – A research design that does not include controls to compensate for the absence of either randomization or a control group.

Probability Sampling – The selection of subjects or sampling units from a population using random procedures; examples include simple random sampling, cluster sampling, and systematic sampling.

Prospective Study – A study that begins with an examination of presumed causes (e.g., cigarette smoking) and then goes forward in time to observe presumed effects (e.g., lung cancer).

Purposive Sampling – A type of nonprobability sampling method in which the researcher selects subjects for the study on the basis of personal judgment about which ones will be most representative or productive; also referred to as judgmental sampling.

Qualitative Data – Information collected in the course of study that is in narrative (nonnumerical) form.

Quantitative Data – Information collected in the course of a study that is in a quantified (numerical) form.

Quasi-Experiment – A study in which subjects cannot be randomly assigned to treatment conditions, although the researcher does manipulate the independent variable and exercises certain controls to enhance the internal validity of the results.

Quota Sampling – The nonrandom selection of subjects in which the researcher prespecifies characteristics of the sample to increase its representativeness.

Random Assignment – The assignment of subjects to treatment conditions in a random manner (i.e., in a manner determined by chance alone); also known as randomization.

Random Sampling – The selection of a sample such that each member of a population has an equal probability of being included.

Randomization – The assignment of subjects to treatment conditions in a random manner (i.e., in a manner determined by chance alone); also known as random assignment.

Range – A measure of variability, consisting of the difference between the highest and lowest values in a distribution of scores.

Reactivity – A measurement distortion arising from the subject’s awareness of being observed or, more generally, from the effect of the measurement procedure itself.

Relationship – A bond or a connection between two or more variables.

Reliability – The degree of consistency or accuracy with which an instrument measures the attribute it is designed to measure.

Replication – The duplication of research procedures in a second investigation for the purpose of determining if earlier results can be repeated.

Representative Sample – A sample whose characteristics are highly similar to that of the population from which it is drawn.

Response Set Bias – The measurement error introduced by the tendency of some individuals to respond to items in characteristic ways (e.g., always agreeing), regardless of the item’s content.

Retrospective Study – A study that begins with manifestation of the dependent variable in the present (e.g., lung cancer) and then links to some presumed cause occurring in the past (e.g., cigarette smoking).

Sample – A subset of a population selected to participate in a research study.

Sample Size – Generally, the greater the number of independent and extraneous variables, the larger the sample should be; usually 10 subjects per each variable is acceptable. A larger sample is likely to be more representative of the population.

Sampling – The process of selecting a portion of the population to represent the entire population.

Sampling Bias – Distortions that arise from the selection of a sample that is not representative of the population from which it was drawn.

Scientific Merit – The degree to which a study is methodologically and conceptually sound and possesses theoretical relevance and internal and external validity.

Secondary Source – A description of a study or studies prepared by someone other than the original researcher.

Selection Bias – A threat to the internal validity of the study that results from differences between groups being compared that can account for differences in the groups’ performances on the dependent variable.

Sensitivity – The probability that a person with a disease will be correctly identified by a clinical test.

Significance Level – The probability that an observed relationship could be caused by chance (i.e., because of sampling error); significance at the 0.05 level indicates the probability that a relationship of the observed magnitude would be found by chance only 5 times out of 100.

Simple Random Sampling – The most basic type of probability sampling, wherein a sampling frame is created by enumerating all members of a population of interest, and then selecting a sample from the sampling frame through completely random procedures.

Specificity – The probability that a person who does not have a disease will be correctly identified by a clinical test.

Standard Deviation – An approximate average of the amount by which a value deviates from the mean.

Standard Error – An estimate of the standard deviation between the estimated value and the true value.

Statistical Significance – A term indicating that the results obtained in an analysis of sample data are unlikely to have been caused by chance at some specified level of probability, generally accepted as < 0.05.

Strata – Subdivisions of the population according to some characteristic (e.g., males and females).

Stratified Random Sampling – The random selection of subjects from two or more strata of the population independently chosen.

Survey Research – A type of nonexperimental research that focuses on obtaining information regarding the status quo of some situation, often through direct questioning of a sample of respondents.

Systematic Sampling – The selection of subjects so that every Xth (e.g., every tenth) person in a sampling frame or list is chosen.

Target Population – The entire population in which the researcher is interested and to which he or she would like to generalize the results of a study.

Test-Retest Reliability – Assessment of the stability of an instrument by correlating the scores obtained on repeated administrations.

Time Sampling – In observational research, the selection of time periods during which observations will take place.

Time Series Design – A quasi-experimental design that involves the collection of information over an extended period of time, with multiple data collection points both before and after the introduction of a treatment.

Treatment – A term used to refer to an experimental intervention or manipulation.

Trend Study – A form of longitudinal study in which different samples from a population are studied over time with respect to some phenomenon.

Type I Error – A decision to reject the null hypothesis when it is true (i.e., the researcher concludes that a relationship exists when, in fact, it does not exist).

Type II Error – A decision to accept the null hypothesis when it is false (i.e., the researcher concludes that no relationship exists when, in fact, it does exist).

Validity – The degree to which an instrument measures what it is intended to measure.

Variable – A characteristic or attribute of a person or object that varies (i.e., takes on different values) within the population under study.

Vulnerable Subjects – Subjects or potential subjects, who may not meaningfully be able to assess the risks of participating in a study and, therefore, may be unable to give voluntary, informed consent (e.g., young children, the mentally retarded, unconscious patients).