A
- alpha – the significance criterion in tests of statistical significance.
- alternate hypothesis – a hypothesis that is different than the null hypothesis.
- analysis of covariance (ANCOVA) – testing the mean group differences on a dependent variable while controlling 1+ covariates.
- analysis of variance (ANOVA) – testing the mean differences among 3 or more groups by comparing the variability between the groups to within the groups.
- anonymity – the participant’s confidentiality is protected from the researcher and publication.
- applied research – research that is intended to find a solution to a practical problem.
- attrition – loss of participants over the course of a study.
B
- between-subjects design – compares separate groups of people.
- bias – any factor that can distort the results of a study and weaken its validity.
- blinding – preventing those involved in the study from having information that could lead to a potential bias.
C
- case-control design – a nonexperimental research design that involves the comparison of a case, e.g., a participant with a condition, e.g., diabetes, and a matched control, e.g., another participant without the condition.
- causal relationship – a relationship between two variables in that one variable determines the other, e.g., cause-effect.
- ceiling effect – the existence of scores near the highest possible value; reduces variability.
- central tendency – the typical set of scores from the center score distribution.
- chi-square test – used to assess differences in proportions.
- coding – the process of transforming raw data into a standardized form for data analysis; either attaching numbers to categories (quantitative) or themes to data (qualitative).
- confidence interval (CI) – range of values within which a population parameter is estimated to be in, e.g., 95% CI.
- confounding variable – a variable that is extraneous to the research question and ‘confounds the independent and dependent variables.
- construct – a concept that is created by researchers for a purpose.
- construct validity – validity from observed persons, settings, and interventions of a study to the constructs.
- content validity – the degree to which items adequately represent the whole of the concept being measured.
- control group – the participants who do not receive the experimental treatment
- correlation – the association between variables.
- covariate – a variable that is held constant in multivariate analyses.
- credibility – used to evaluate integrity and quality in qualitative studies (internal validity in quantitative research).
D
- degrees of freedom (df) – the number of sample values that are free to vary within a sample mean.
- dependability – stability of qualitative data over time and condition (reliability in quantitative research).
- dependent variable – the variable that depends on or is caused by another variable (independent); outcome variable.
- double-blind study – a study in which two groups are blinded and neither the subjects nor researchers know which group is which.
E
- effect size – the magnitude of the relationship between two variables.
- ethnography – the study of the culture of a group of people.
- experiment – a study where the researcher controls the independent variable and assigns subjects randomly to different treatment conditions; RCTs.
- external validity – the degree to which study results can be generalized to ones other than the one studied.
- extraneous variable – a variable that confounds with the independent and dependent variables.
F
- face validity – the extent to which a measuring tool looks as though it is measuring what it intends to measure.
- feasibility study – a smaller-scale study to assess the viability of a future larger study; a pilot study.
G
- generalizability – findings are applicable to a broader group than study participants.
- grounded theory – collecting qualitative data to develop theories grounded in real-world observations.
H
- Hawthrone effect – an effect on the dependent variable that results when participants are aware they are being studied.
- homogeneity – the reliability in which the parts are internally consistent or similar.
- hypothesis – a statement that predicts a relationship between variables within a population.
I
- independent variable – the variable that is believed to cause influence on the dependent variable; manipulated variable.
- inference – drawing a conclusion from the study’s evidence.
- informed consent – (ethics) requires obtaining voluntary permission and participation of participants after informing them of the risks and benefits.
- instrument – something used to collect data.
- intercoder reliability – the degree to which coders agree on coding decisions.
- internal validity – the degree to which the independent variable is responsible for the effects, and not extraneous factors.
- interrater reliability – the degree two to which two observers (independently) assign the same value for the measurement.
- intervention fidelity – the degree to which the implementation of the treatment is faithful to its original intention.
- investigator triangulation – using multiple researchers to analyze data; enhances rigor.
L
- level of significance – the risk of making a Type I error with the established alpha.
- Likert scale – a measure involving a set of items that respondents rate their degree of agreement vs disagreement.
- longitudinal study – data collection over a period of time; contrasts cross-sectional studies.
M
- maturation threat – a threat to internal validity due to the passage of time.
- mean – a central tendency measure that sums the scores and divides them by the total number.
- meta-analysis – studies the results from multiple similar studies and research questions.
- mixed methods research – both qualitative and quantitative data are collected and analyzed.
- mode – a measure of central tendency in which the value occurs most frequently in the distribution of scores.
- mortality threat – loss of participants; threat to internal validity.
- multivariate analysis of variance (MANOVA) – tests the significance of differences between the means of 2+ groups on 2+ dependent variables.
N
- N – the total number of participants (n = subgroups within a study)
- needs assessment – determines the needs of a group such as a community or an organization used for planning or allocation.
- negative relationship – as one variable increases, the other decreases; inverse relationship.
- nonexperimental research – observational research
- nonprobability sampling – using non-random procedures for sampling, e.g., convenience sampling.
- normal distribution – a bell-shaped curve; normal curve.
- null hypothesis – predicts no relationship between the variables; used as the hypothesis to be rejected.
- one-tailed test – only values of one side (tail) of a distribution are considered to be significant
- ordinal measurement – “order”, ranking phenomenon along a dimension.
P
- p-value – the probability that the results are due to chance alone; probability of a type I error.
- Pearson’s r – correlation coefficient of the magnitude of the relationship between two variables.
- phenomenology – the study of the lived experience of humans.
- pilot study – a smaller scale version of a study completed in preparation of a larger study; feasibility study.
- placebo – sham intervention, often used in control groups.
- placebo effect – changes to the dependent variable due to the placebo.
- population – the entire set of people with some common characteristic.
- positive relationship – two variables occur in the same direction, e.g., an increase in one and an increase in another.
- posttest – data that is collected after the introduction of an intervention
- power – if a study’s design can detect relationships among variables.
- pretest – the collection of data prior to the experiment’s intervention is introduced.
- pretest-posttest design – data is collected both before and after the introduction of an intervention.
- principal investigators (PI) – the lead researcher who oversees a study.
- probability sampling – the selection of participants from a population using random procedures
- psychometric assessment – an evaluation of the quality of an instrument based on its reliability and validity.
- psychometrics – the study of the principles and application of theories for developing measuring tools in research.
- publication bias – the tendency for published studies to over-represent the reality of findings; not publishing negative results; bias against the null hypothesis.
Q
- qualitative research – investigation of some phenomenon through the collection of narrative materials, e.g., interview.
- quantitative research – investigation of some phenomenon that leads to precise measurements using quantification (e.g., numbers).
R
- r – Pearson’s r; magnitude and direction of a relationship between two variables.
- R – indicates the magnitude, but not the direction of a relationship between two variables.
- randomized controlled trial (RCT) – a traditional experiment that involves random assignment to treatment groups; phase III of clinical trials.
- range – a measure of variability from highest to lowest distribution of scores.
- regression analysis – used to predict values of a dependent variable based on 1+ independent variables.
- relationship – strength or connection between 2+ variables.
- reliability – the dependability to which an instrument measures its attribute.
- representative sample – whether a sample is comparable to the larger population.
- retrospective design – a study that begins with the present dependent variable followed by finding a cause from the past.
S
- sample – participants that come from a subset of the population.
- sampling bias – when a sample is not representative of the population.
- scoping review – a review of research findings designed to refine the questions for a systematic review.
- setting – where data collection takes place in a study.
- significance level – the probability that a relationship could be due to chance.
- snowball sampling – the selection of participants through a referral system of prior participants; network sampling.
- standard deviation – statistical measure for the degree of variability in a set of scores.
- standard scores – scores that are expressed in standard deviations from the mean; z scores.
- statistical-significance – the results from sample data are unlikely to have been caused by chance.
- strata – subdivisions of the population based on a characteristic, e.g., ethnicity.
- systematic review – a synthesis of research findings that uses systematic sampling and analysis of data.
T
- test-retest reliability – assessment of an instrument by correlating the scores obtained on multiple instances.
- transferability – the extent that which qualitative findings can be transferred to other settings; generalizability.
- triangulation – using multiple methods of collection and interpretation of data to converge on reality.
- trustworthiness – the degree of confidence that which qualitative researchers have in their data.
- t-test – a statistical test for analyzing the difference between two means.
- Type I error – a false positive; rejecting the null hypothesis when it is true.
- Type II error – a false negative; accepting the null hypothesis when it is false.
- two-tailed test – both ends of the sampling distribution are used to determine the probability of values.
V
- validity – the degree to which inferences can be made in a study based on accuracy; if an instrument measures what it intends to measure.
- variability – the dispersal of a set of scores.
- variable – an attribute that can contain different values.
- variance – dispersion; standard deviation squared.
- visual analog scale (VAS) – a scale used to measure a variable on a straight line.
Z
- z-score – a standard score that is expressed in terms of standard deviations from the mean.
References
Child Care & Early Education Research Connections. (n.d.). Research Glossary. Retrieved from https://www.researchconnections.org/research-tools/research-glossary/
OECD. (n.d.). Glossary of Statistical Terms. Retrieved from https://stats.oecd.org/glossary/index.htm
SAGE Research Methods. (2006). The SAGE Dictionary of Social Research Methods. Retrieved from https://methods.sagepub.com/Reference/the-sage-dictionary-of-social-research-methods
The Cochrane Collaboration. (2005). Glossary of Terms in The Cochrane Collaboration. Retrieved from http://aaz.hr/resources/pages/57/7.%20Cochrane%20glossary.pdf
University of Southern California. (2021). Glossary of Research Terms. Retrieved from https://libguides.usc.edu/writingguide/researchglossary