Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.

Normal Size Small Size show me how

Normal Size Small Size show me how

# Clinical Research

### Clinical Research Quiz 2 pt 2

Term | Definition |
---|---|

Cronbach’s alpha/ Coefficient alpha | a reliability index that estimates the internal consistency or homogeneity of a composite measure composed of several items or subparts; DOESN’T evaluate fluctuations over time as a source or unreliability |

Concurrent validity | the degree to which scores on an instrument are correlated with an external criterion, measured at the same time; an instrument’s ability to distinguish individuals who differ on a present criterion |

Difference between predictive and concurrent validity | The difference in timing of obtaining measurements on a criterion |

Confirmability | a criterion for integrity referring to the objectivity or neutrality of the data & interpretations |

Confirmatory factor analysis (CFA) | a factor analysis designed to confirm a hypothesized measurement model, using maximum likelihood estimation |

Construct validity | t/ validity of inferences from observed persons, settings, & interventions in a study to the constructs that these instances might represent; w/ an instrument, t/ degree to which it measures the construct under investigation |

Content validity | t/ degree to which the items in an instrument adequately represent the universe of content for the concept being measured |

Convergence | an approach to construct validation that involves assessing the degree to which 2 methods of measuring a construct yield similar information; evidence that different methods of measuring a construct yield similar results |

Correlation coefficient | a tool used to quantitatively describe the magnitude and direction of a relationship btw 2 variables |

Credibility | in QUALITATIVE research, the confidence in the truth of the data (in QUANTITATIVE research, the like term is internal validity) |

Criterion-related validity | degree to which scores on instrument are correlated w/ external criterion (i.e. if scores on scale of attitudes toward premarital sex correlate highly w/ subsequent loss of virginity in sample of teenagers, then attitude scale would have good validity) |

Cut-off score | a key in developing new diagnostic or screening measures; uses a score to distinguish cases & noncases |

Data triangulation | use of multiple data sources for the purpose of validating conclusions |

Dependability | in QUALITATIVE research, the stability of the data over time & conditions; (in QUANTITATIVE research it’s reliability) |

Discriminability | the ability to differentiate the construct form other similar constructs |

Equivalence | degree to which two or more independent observers or coders agree about scoring; high agreement= measurement errors hv bn minimized |

Error of measurement | the difference between true and obtained scores that is a result of factors that distort the measurement |

Face validity | refers to whether t/ instrument appears, on the face of it, to be measuring the appropriate construct |

Factor | a dimension that represents a relatively unitary attribute |

Factor analysis | method for identifying clusters of related variables |

Internal consistency | t/ degree to which the subparts of a composite scale are all measuring the same attribute or dimension, as a measure of the scale’s reliability |

Interrater (interobserver) reliability | t/ degree to which 2 raters/observers, operating independently, assign the same ratings or values for an attribute being measured or observed |

Inverse relationship | incr in one variable are associated with decr in the 2nd variable. |

Investigator triangulation | use of 2 or more researchers to analyze and interpret a data set, to enhance rigor |

Item analysis techniques | type of analysis used to assess whether items on a scale are tapping the same construct and are sufficiently discriminating |

Item response theory (IRT)/ latent trait theory | a measurement perspective where the focus is on the item rather than the overall scale or rests, and procedures involve examining a person’s response to each item |

Known-groups technique | technique for estimating the construct validity of an instrument through an analysis of the degree to which the instrument separate groups predicted to differ based on known characteristics or theory |

Measurement | involves assigning numbers to represent the amount of an attribute present in an object or person, using a specified set of rules |

Member check | method of validating the credibility of qualitative data through debriefings & discussions w/ informants |

Method triangulation | use of multiple methods of data collection about the same phenomenon, to enhance rigor or validity |

Multitrait-multimethod matrix method (MTMM) | using multi measures to assess instrument’s construct validity; instrument valid due to relation btw it & other measures of same attribute (convergence) & a weak relationship btw it & measures purporting to measure diff attribute (discriminability) |

Observed (obtained) score | w/in classical measurement theory, the actual score/numerical value assigned to a person on a measure; obtained score can be conceptualized as having two parts—an error component and a true component. |

How is obtained score calculated? | Obtained score = True score + Error |

Peer debriefing | sessions w/ peers to review & explore various aspects of a study, sometimes used to enhance trustworthiness in a qualitative study |

Perfect relationship | a correlation btw 2 variables such that the values of one variable permit perfect prediction of the values of the other; designated as 1 or -1 (i.e. assuming that tallest person is heaviest person, 2nd tallest is 2nd heaviest, etc.) |

Persistent observation | a QUALITATIVE researcher’s intense focus on the aspects of a situation that are relevant to the phenomena being studied |

Person triangulation | collection of data from diff levels of persons, w/ the aim of validating data through multiple perspectives on the phenomenon |

Positive relationship | relationship btw 2 variables where high values on one variable tend to be associated with high values on the other (i.e. as phys activity incr, hrt rate incr too) |

Predictive validity | t/ adequacy of an instrument in differentiating btw people’s performance on a future criterion (i.e. school correlates high school students’ grades /w performance in college) |

Prolonged engagement | in QUALITATIVE research, the investment of sufficient time during data collection to have an in-depth understanding of the group or phenomenon under study, thereby enhancing credibility |

Psychometric assessment | an evaluation of the quality of an instrument, based primarily on evidence of its reliability and validity |

Reliability | accuracy and consistency of information obtained in a study; the degree of consistency or dependability /c which an instrument measures an attribute |

Reliability coefficient | an index of the magnitude of the test’s reliability; this allows test-retest reliability comparison to be performed objectively |

Researcher credibility | the faith that can be put in a researcher, based on his or her training, qualifications, and experience |

Sensitivity | the ability of screening instruments to correctly identify a “case” that is, to CORRECTLY DIAGNOSE A CONDITION |

Specificity | t/ ability of screening instrument to CORRECTLY IDENTIFY NONCASES |

Stability | t/ extent to which similar scores are obtained on separate occasions |

Test-retest reliability | t/ assessment of the STABILITY of an instrument by correlating the scores obtained on 2 administrations |

Theory triangulation | use of competing theories or hypotheses in the analysis and interpretation of data |

Thick description | a rich and thorough description of the research context in a qualitative study |

Time triangulation | the collection of data on the same phenomenon or about the same people at diff points in time, to enhance validity |

Transferability | in QUALITATIVE RESEARCH the extent to which qualitative findings can be transferred to other settings; (in QUANTITATIVE research, it’s called generalizability) |

Triangulation | the use of multiple sources or references to draw conclusions about what constitutes the truth; people combine qualitative interviews /c quantitative study tools |

True score | a hypothetical score that would be obtained if a measure were infallible |

Trustworthiness of data | the degree of confidence QUALITATIVE researchers have in their data; it’s assessed using credibility, transferability, dependability, confirmability, & authenticity |

Validity | complex concept that broadly concerns the SOUNDNESS of a study’s evidence—that is, whether the findings are unbiased & well grounded; the degree to WHICH AN INSTRUMENT MEASURES WHAT IT’S INTENDED TO MEASURE |

Clarity | scale items should be clear, unambiguous items; words should be carefully chosen with the educational & reading level of the target population in mind; to be considered while wording items for scale |

Confirmatory factor analysis (CFA) | a factor analysis designed to confirm a hypothesized measurement model, using maximum likelihood estimation |

Construct | an abstraction or concept that is deliberately invented by researchers for a scientific purpose (i.e. health locus of control) |

Content validation | t/ process in which the items in an instrument adequately represent the universe of content for the concept being measured |

Cut-off point (score) | a key in developing new diagnostic or screening measures; uses a score to distinguish cases & noncases |

Difficulty levels | an item analysis technique which indicates how “difficult” each item is |

Domain sampling model | Random sampling of a homogeneous set of items from a hypothetical universe of items relating to the construct |

Eigenvalues | in FACTOR ANALYSIS the value equal to the sum of the squared weights for each factor |

Exploratory factor analysis (EFA) | A FACTOR ANALYSIS undertaken to explore the underlying dimensionality of a set of variables |

Factor analysis | method for identifying clusters of related variables |

Factor extraction | 1st PHASE of FACTOR ANALYSIS, involving extraction of as much variance as pos thru t/ successive creation of linear combinations of variables in data set; condensing items to smaller no. of factors & used to identify t/ no. or underlying dimensions |

Factor loadings | in FACTOR ANALYSIS, t/ weight associated w/ variable on a given factor |

Factor rotation | t/ 2ND PHASE OF FACTOR ANALYSIS, during which, the reference axes for the factors are moved to more clearly align variables w/ a factor; makes factors more interpretable |

Goodness-of-fit statistic | Testing hypothesized measurement model against actual data, showing correlation & indicating if model is a good fit |

Item analysis | a type of analysis used to assess whether items on a scale are tapping the same construct and are sufficiently discriminating |

Jargon | terms that might be well-known in one field of existence, but not in another; should be avoided in forming scale items |

Latent variable | underlying construct; an unmeasured variable that represents an underlying, abstract construct |

Maximum likelihood estimation | an estimation approach in which the estimators are ones that estimate the parameters more likely to have generated the observed measurements |

Measurement model | in structural equations modeling, the model that stipulates the hypothesized relationships among the manifest and latent variables |

Oblique rotation | in FACTOR ANALYSIS, a rotation of factors such that the reference axes are allowed to move to acute or oblique angles and hence the factors are allowed to be correlated; greater than 90-degrees; allows correlated factors |

Orthogonal rotation | in FACTOR ANALYSIS, a rotation of factors such that the reference axes are kept at right angles, and hence the factors remain uncorrelated; maintains independence of the factors |

Principal components analysis (PCA) | a widely used FACTOR extractor method where weights for the first factor are compute such that the avg squared weight is maximized |

Readability | t/ ease with which materials can be read by people w/ varying reading skills, often empirically evaluated through readability formulas |

Created by:
Fukanwa