Chapter 6 Flashcards
The probability of a Type I error. It is selected during the design of the research and represents the researchers’ risk of incorrectly rejecting the null hypothesis (of saying there is a difference when in fact there is not).
alpha
The risk of a Type I error when a test of equality is performed. If equality is tested the researchers assume that either group could turn out to be superior, in which case alpha should be divided in half.
alpha-halves
The ratio of the false positive rate to the true positive rate produces a proportion called the area under the curve.
AUC (area under the curve)
The square of r (Pearson’s product-moment correlation coefficient). It tells us the proportion of the variability in the response variable that is explained by the variability in the predictor variable.
coefficient of determination
Selecting the subjects of a study based on ease of access (convenience).
convenience sampling
The number of observations minus 1.
df (degrees of freedom)
In a statistical power calculation, effect size is the expected amount of effect that is considered a minimal clinically meaningful effect.
effect size
A procedure for combining items in a survey instrument to represent a single concept (called a construct or domain). A survey is broken down into domains (also called subscales) that include multiple questions.
factor analysis
A type of chi-square test that can test if the sample distribution fits with a theoretical distribution.
goodness-of-fit test
A procedure in which sampling occurs in one or more steps (stages). Often this procedure is used when dealing with very large sampling frames and researchers want to ensure that every subgroup is included
multistage sampling
A procedure in which subjects are purposely selected in order to ensure that they represent/experience the phenomenon of interest. This is often used with rare conditions or with small populations.
nonprobability sampling
A quantitative test that compares the mean score of a sample to the known mean (normative or expected) value from a trustworthy source.
one-sample t-test
A statistical procedure to determine the probability that the effect of interest will be observed when it occurs. A power level of 0.8 is generally accepted as a sufficient level of power.
power calculation
Typically used in qualitative research studies, such as grounded theory, phenomenology, ethnography, or biography.
purposive sampling
Test that focuses on relationships among several variables that best explain a specific response.
regression analysis
The consistency, or reproducibility, of the manner in which the data were collected.
reliability
Pre/post, within-subject, self-controlled, and correlated group designs are all types of these.
repeated measures tests
The proportion that a sample is of the population it represents.
sampling fraction
A form of probability sampling (which means each individual has a known probability of being selected) that allows every member of the population to have an equal chance of being selected for the study.
simple random sample
A type of nonprobability sampling used in quantitative research when the researchers are interested in studying a population that is difficult to identify, such as gang members, IV drug users, or illegal aliens, to name a few.
snowball sampling
A procedure in which the sampling frame is divided into strata (groupings) and a random sample is drawn from each stratum
stratified random sample
Comparison of a statistic, such as a mean or a proportion, between two or more groups.
tests for differences
Used to determine if variables change in relationship to other variables (e.g., arthritis incidence increases with age or heart disease increases with incidence of diabetes).
tests for relationships
The accuracy of measurements. It is the degree to which an instrument measures what is it supposed to measure.
validity
