Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
40 Cards in this Set
- Front
- Back
Validity
|
-refers to whether or not the instrument accurately measures what it is suppose to measure
|
|
Content Validity
|
-represents the universe content or domain (population) of a given construct
-subtype is face validity: basically verifies that the instrument gives the appearance of measuring the concept |
|
Criterion-related Validity
|
-indicates to what degree the subject’s performance on the measurement tool and the subject’s actual behavior are related
|
|
Concurrent:
|
the degree of correlation of two measures of the same concept being measured at the same time.
Ex. SAT scores compared to the high school record |
|
Predictive:
|
the degree of correlation of the concept and some future measure of the same concept
Ex. SAT scores predict how well one will do in college |
|
Construct Validity
|
-based on the extent to which a test measures a theoretical construct or trait
-establishment of construct validity is a complex process often involving several studies and approaches that include |
|
• Hypothesis Testing Approach
|
investigator uses the theory or concept underlying the instrument to validate the instrument
|
|
• Convergent validity
|
refers to a search for other measures of the construct
|
|
• Divergent validity
|
uses measurement approaches that differentiates one construct from another that may be similar or opposite of each other
|
|
• Multitrait-multimethod
|
combines both convergent and divergent
|
|
• Contrasted-Groups Approach
|
the researcher identifies 2 groups of individuals who are expected to score very high or extremely low in the characteristics being measured by the instrument
|
|
Who rose as superpowers after WWII?
|
US and USSR
|
|
Threats to Internal Validity (did the independent variable cause the change in the dependent variable)
|
1. History
• what was going on during the time of the study • Sometimes environment has an affect on the outcomes of the study 2. Maturation • The participants mature and behavior changes with maturity 3. Testing • Person may recognize things that were on the pretest that are the same as the posttest 4. Instrumentation • Was the instruments calibrated right? If not can influence the study’s outcomes 5. Mortality • Lost of participants 6. Selection Bias • Pool of individuals may be similar to each other without knowing it |
|
Reliability
|
Refers to the ability of the instrument to produce the same result on different occasions
Includes: • Stability of an instrument: where the instrument is able to produce the same results with repeated measures • Homogeneity of an instrument: where the instrument is measuring all the same concepts or characteristics • Equivalence is the tool produces the same results when equivalent or parallel procedures are used -use this to make sure we are reducing error |
|
• Reliability Coefficient Interpretation
|
expresses the relationship among the error variance, true variance and the observed score and ranges from 0-1 (0.07 is acceptable in research studies)
Test used to calculate reliable coefficient (Way to assure results are because of the intervention and not by chance) |
|
Test-retest
|
administration of the same instrument to same subjects under similar conditions on 2 or more occasions
|
|
Parallel or alternate form
|
same individuals tested within a specific interval but a different form of same test is given to the subjects on 2nd testing
|
|
Split haf
|
involves dividing the scale in half and making comparisons
|
|
Kuder-Richardson
|
used for instruments that have a dichotomous response format-yields correlation based on consistency of responses to all items of a single form of a test that is administered one time
|
|
Interrater reliability:
|
used in direct measurements of observed behavior
-2 or more individuals make observations on several occasions and compare findings -expressed as a percentage of agreement between the observers or as a correlation coefficient of the scores assigned to the observed behaviors |
|
Levels of Measurement 1. Nominal
|
lowest of the four measurement categories
-used when the data can be organized into categories of a defined property but the categories cannot be compared -categories must not be orderable Ex. Gender, ethnicity, martial status and diagnosis |
|
Levels of Measurement 2. Ordinal
|
-data are assigned to categories that can be ranked
-one category is judged to be or ranked higher than another -considered to have unequal intervals |
|
Levels of Measurement3. Interval Scale
|
-uses interval scales which have equal numerical distances between the intervals
-represents a continuum of values -magnitude of attribute can be more precisely defined -lacks a zero point Ex. Temperature is most common interval scale |
|
Levels of Measurement4. Ratio Scale
|
highest form of measurement and meets all of the other forms of measurement: mutually exclusive, exhaustive, ordered ranks, equally spaced intervals and continuum values
-have absolute zero points and means absence of the property Ex. Weight, length, volume |
|
Descriptive Statistics
|
• Used in any study where the data uses numbers
• Also known as summary statistics • Summary statistics and visual displays that illustrate the characteristics of a study sample • Includes measures of central tendency • Can also be used to describe differences between groups or variables |
|
• Frequency
|
involves ordering the numbers from lowest to highest and counting the # of times each value appears in the data
|
|
Inferential Statistics
|
goes beyond describing the sample
-allows the researcher to generalize from the sample to the larger population -based on the assumption that the sample is randomly selected Sample vs. Population Sample: subset of the population Population: is the entire set of elements that the researcher wants to study |
|
Statistical Significance
|
-refers to that there is a relationship between the variables and that the results were not because of chance
-just because there is statistical significance does not mean there is clinical significance |
|
Clinical Significance
|
-refers to does the statistical significance has relevance to practice
-is the findings big enough to change the practice |
|
Nonsignificance
|
-refers to that any observed difference or relationships could have been the result of chance fluctuations
|
|
Ways of Testing Statistical Significance (causality)
|
Causality is a way of knowing that one thing causes the other
-They can be used to understand the effect of an intervention and are critical to the development of nursing science -These statistic examine causality by examining differences between and among groups |
|
• T-test for independent Samples
|
Used when the researcher is interested in comparing 2 independent groups of subjects with respect to the average values on a dependent variable
-Interval level data is important for this test -When the t-test is used a t- value is computed and compared to the critical value in the table and if the t-value is greater than or equal to the value in the table the groups are significantly difference -If the data is nominal or ordinal a chi-square test can be used: which examines is the variables related or independent from one another |
|
• Paired T-test
|
-Used when the researcher wants to compare 2 sets of mean scores from the same subjects
|
|
• Analysis of Variance
|
procedure used to test significant differences between averages
-use of 3 or more mean scores -reported as a F-statitistic and if the computed F-value is greater than or equal to the F-value in statistical table then there is a significant difference between the groups Ex. The researcher is interested and comparing 3 interventions to help people stop smoking |
|
Probability
|
-requires that every element in a population have an equal chance of being selected
-helps to have a representative sample |
|
• Simple Random Sampling
|
-each element of the population have an equal chance of inclusion into the sample
-assign each subject a number and close eyes and point to numbers randomly until sample size is reaches |
|
• Stratified Sample
|
-population is divided into strata
-sample a number of people from each strata |
|
• Systematic Sampling
|
-involves drawing from every nth element in a population
Ex. Drawing every 10th element in a population |
|
• Cluster Sampling
|
-requires that the population be divided into clusters or groups
-most large scale studies use this |
|
Nonprobability
|
-there is no way of knowing that the probability of each element will be included in the sample
-results are NOT generalized to the population |