ch 15
Which type of validity is demonstrated when a nurse researcher submits items of an instrument to a panel of experts to evaluate item clarity? A. Content validity B. Construct validity C. Concurrent validity D. Criterion-related validity
ANS: A A panel of experts judges instrument validity through the content type of validity. Construct validity validates the theory underlying the measurement and testing of hypothesized relationships. Concurrent validity refers to the degree of correlation of two measures of the same concept administered at the same time. Criterion-related validity incorporates two forms: concurrent and predictive.
What property is established when the results of two separately administered tests that measure the same domain or concept are highly correlated? a. Validity b. Stability c. Equivalence d. Homogeneity
ANS: A Convergent validity example: two or more instruments that theoretically measure the same construct and are administered to the same subjects. The example in the stem matches validity, not reliability. The example in the stem matches validity, not reliability. The example in the stem matches validity, not reliability
A nurse researcher determines an instrument's homogeneity using which type of reliability? Split-half reliability Test-retest reliability Interrater reliability Alternate form reliability
ANS: A Instrument homogeneity is established with split-half reliability. Test-retest reliability establishes instrument stability. Interrater reliability is not a homogeneity measure. Alternate form or parallel form reliability is not a test of homogeneity.
A nurse researcher should be concerned about an instrument's degree of internal consistency because it means: a. the instrument is appropriate to use to measure a single concept. b. the instrument has low measurement error and high error variance. c. more refinement of the instrument is needed before it can be applied. d. the instrument is valid, but the reliability has yet to be determined.
ANS: A Internal consistency or homogeneity reliability indicates that the items within the scale measure the same concept .This is an inaccurate statement. A high degree of internal consistency indicates that the instrument is refined; however, additional testing is always warranted. There is no indication that the instrument is valid; the reliability coefficient is very acceptable.
Testing of a new instrument demonstrates that it has a high degree of internal consistency. To the nurse researcher this indicates: a. the instrument is appropriate to use to measure a single concept. b. the instrument has low measurement error and high error variance. c. more refinement of the instrument is necessary before it can be applied. d. the instrument is valid, but the reliability has yet to be determined.
ANS: A Internal consistency or homogeneity reliability indicates that the items within the scale measure the same concept. This is an inaccurate statement. A high degree of internal consistency indicates that the instrument is refined; however, additional testing is always warranted. There is no indication that the instrument is valid; the reliability coefficient is very acceptable.
A nurse researcher assesses the degree to which individual items on a scale cluster together around multiple dimensions using what? Factor analysis Split-half reliability Cronbach's alpha coefficient Kuder-Richardson coefficient
ANS: A The extent to which a set of items measures the same underlying construct or dimension of a construct and clusters together around one or more dimensions describes factor analytic construct validity. This reliability is not associated with factor analysis and construct validity. This reliability is not associated with factor analysis and construct validity. This reliability is not associated with factor analysis and construct validity
An ear temperature probe that consistently reports body temperature at a degree lower than the patient's actual temperature has what type of reliability or validity problem? Reduced reliability, systematic error Reduced validity, random error Increased validity, systematic error Increased validity, random error
ANS: A The instrument is systematically measuring temperature inaccurately. This response does not refer to random error. The validity of the instrument is not increased. The validity of the instrument is not increased.
Concurrent validity indicates: A. the degree of correlation between the measure of the concept and some future measure of the same concept. B. the degree of correlation of two measures of the same concept administered at the same time. C. the extent to which a test measures a theoretical construct or trait. D. the representativeness of the items to measure a specific concept.
ANS: B Concurrent validity is established when two measures of the same construct are administered at the same time, not a future measure. This response fits the definition of concurrent validity. This answer fits the description of construct validity. This response fits the description of content validity.
Which type of validity is most difficult to establish? Content validity Construct validity Predictive validity Concurrent validity
ANS: B Face and expert validity are relatively easy to establish. Construct validity includes many types of validity; it is complex and tends to take much more time to establish than other types of validity. Predictive validity is relatively easy to establish. Concurrent validity is relatively easy to establish.
A nurse researcher would want to use a Kuder-Richardson (KR-20) coefficient to establish the internal consistency of an instrument: when questions are open-ended. when multiple choice tests are analyzed. when the instrument uses a Likert-type response scale. when the instrument is designed to measure more than one concept.
ANS: B Open-ended questions elicit textual data that are content analyzed (qualitative analysis). The KR-20 coefficient provides estimates of homogeneity used for instruments that have a dichotomous response format. Cronbach's alpha coefficient provides estimates of homogeneity when instruments are scaled with ordinal responses. The KR-20 coefficient does not apply to this answer.
The nurse researcher notes that the initial test for reliability of an instrument has been conducted on a sample that has different characteristics from those of the current study sample. The researcher should: a. conduct a pilot study on the current sample to determine whether reliability is maintained. b. use the instrument with the current sample because reliability holds across samples. c. use the Kuder-Richardson formula to recalculate the reliability coefficient. d. discard or reject the instrument.
ANS: B The pilot sample was most likely done before this time. A reliable instrument can be used on more than one occasion to measure a set of behaviors that ordinarily remains relatively constant. Kuder-Richardson could be calculated on the current sample, but this answer is incorrect. The instrument should not be discarded.
A nurse researcher describes an instrument that is administered repeatedly and obtains the same results as demonstrating: a. validity. b. reliability. c. consistency. d. predictability.
ANS: B The stem indicates reliability, not validity. This answer indicates consistency and stability of an instrument. Consistency is a characteristic of reliability. Predictability is a characteristic of reliability.
What type of validity is demonstrated by examining the extent to which a passing grade in a nursing course is correlated with passing the registered nurse licensure examination? Content validity Construct validity Predictive validity Concurrent validity
ANS: C Content validity refers to the representativeness of items to measure a specific concept. Construct validity refers to the extent to which a test measures a theoretical construct or trait. This criterion-related validity example is predictive and connects past with future performance. The example in the stem does not fit concurrent validity.
A nurse researcher would want to use a Cronbach's alpha coefficient to establish the internal consistency of an instrument: when questions are open-ended. when questions/statements demand a yes or no response. when the instrument uses a Likert-type response scale. when the instrument is designed to measure more than one concept.
ANS: C Open-ended questions elicit textual data that are content analyzed (qualitative analysis). KR-20 provides estimates of homogeneity used for instruments that have a dichotomous response format. Cronbach's alpha coefficient provides estimates of homogeneity when instruments are scaled with ordinal responses. Cronbach's alpha coefficient does not apply to this answer.
A nurse researcher administered a test anxiety questionnaire to a group of nursing students before they attended a test anxiety workshop and then administered a different questionnaire to the same nurses immediately following the workshop. This is indicative of which type of reliability? Split-half Test-retest Parallel form Alternate form
ANS: C Split-half refers to internal consistency and involves dividing a scale into two halves and making a comparison. This response fits a quantitative study only partially and is incomplete for a quantitative, quasi-experimental study. Parallel form means that the same individuals are tested within a specific interval but the subjects are given a different form of the test on the second testing. This response fits with quantitative tests.
A nurse researcher describes an instrument that produces the same result when it is administered to the same subjects under similar conditions on two or more occasions as possessing: a. homogeneity. b. equivalence. c. stability. d. validity.
ANS: C The instrument is administered to the same subjects under similar conditions on two or more occasions. Correlations are strong. Test-retest reliability establishes stability of the instrument. Parallel forms is a reliability example. The same results are obtained on repeated administration of the instrument. This stem is a reliability example, not a validity example.
Nurse researchers critiquing research reports should be concerned with the assessment of the validity and reliability of study instruments: a. to determine the utility of the instruments for triangulation. b. to assess the relationships between the hypotheses and the research questions. c. to determine whether the concepts and variables were measured adequately. d. to assess whether the concept under study is being treated as a dependent or an independent variable.
ANS: C There are several types of triangulation; the instrument chosen might assist a study to become triangulated, but the reliability and validity of it must be determined before triangulation. This response fits the connection between hypotheses and research questions but not instrument reliability and validity. An indicator of a study's excellence is the establishment of the reliability and validity of the instruments used to measure variables. Instrument reliability and validity issues are separate issues from whether variables are independent or dependent variables.
A new instrument has been determined to be highly valid. The nurse researcher should interpret this finding to mean: it is sensitive but not specific. its use results in minimal random errors. it accurately measures level of mentation. determination of interrater reliability is unnecessary.
ANS: C This answer does not reflect validity, but rather it reflects characteristics of screening tests. Validity does not guarantee minimal random errors. Validity of an instrument is associated with its accuracy. Interrater reliability is not an example of validity.
The nurse researcher notes that test-retest correlations were r = 0.79 when given over 4-week intervals. This is interpreted to indicate: a. equivalence. b. discriminability. c. reliability. d. homogeneity.
ANS: C This response is one type of reliability. This response refers to a type of validity. The example of test-retest correlations fits reliability. Test-retest procedures establish stability of instruments, not homogeneity.
The reliability coefficient of a new instrument is established at 0.86. The nurse researcher should interpret this finding as: high error variance; high reliability. high error variance; low reliability. low error variance; high reliability. low error variance; low reliability.
ANS: C When the error variance is low, the reliability coefficient will be closer to 1; a reliability level of 0.70 or higher is considered acceptable. When the error variance is low, the reliability coefficient will be closer to 1; a reliability level of 0.70 or higher is considered acceptable. When the error variance is low, the reliability coefficient will be closer to 1; a reliability level of 0.70 or higher is considered acceptable. When the error variance is low, the reliability coefficient will be closer to 1; a reliability level of 0.70 or higher is considered acceptable.
When the nurse researcher demonstrates that an instrument is highly reliable, which type of error is reduced? Random error Variance error Persistent error Systematic error
ANS: D Random error is connected with validity. Variance and error are connected as a research term, error variance, not the reverse; it is defined as variability in test scores. Persistent error is not a research term. Systematic error is connected with reliability
Which type of reliability exists when consistency of observation, measured by a correlation coefficient, is noted between two or more research assistants who record their observations of an event at the same time? Internal consistency Parallel forms Test-retest Interrater
ANS: D The example illustrates a test of homogeneity. The example does not include two instruments measuring the same construct. The example does not include a time interval between administration of the instrument. The consistency of observations between two raters making an observation illustrates interrater reliability.
During development of an instrument to measure self-esteem, the nurse researcher administered the instrument to individuals who were substance abusers and to individuals who were not substance abusers and anticipated a significant difference in scores. This method of establishing construct validity is categorized as what? Factor analysis Convergent validity Discriminant validity Contrasted-groups approach
ANS: D The example in the stem does not provide an example of construct validity using factor analysis. The example does not illustrate convergent validity. The example does not illustrate discriminant validity. Two groups of subjects who are suspected to score extremely high or low in the characteristic being measured complete the instrument and differences in scores are examined; the groups should differ and construct validity will be supported.
A researcher who is developing a new instrument to measure pain has been informed that the instrument has face validity. The researcher's next step should be to: use the instrument in a parent study. use the instrument in a pilot study. assess the reliability of the instrument. assess the content validity of the instrument.
ANS: D The findings would be in question if this were the next step. Another step should be taken before testing is done. Reliability testing is not the next step; establishing greater validity should occur first. The next step is content validity established through literature and expert judges
In testing an instrument consisting of 25 items for homogeneity using the "item-to-total" correlation, 8 items were found to have a low correlation to the total. The nurse researcher interprets this information to mean that the researcher should: a. use the instrument without changes. b. use the instrument only with a multitrait-multimethod approach. c. retain the 8 items with low correlation and delete the other 17 items. d. delete the 8 items with low correlation and retain the other 17 items.
ANS: D The instrument should be modified. Items should be deleted. The multitrait-multimethod approach is appropriate for ongoing instrument development and validity. The reverse of this answer should be performed: 17 items should be retained and 8 items should be eliminated. Eight items should be deleted and 17 items should be retained.
A group of researchers developing an instrument performed a test-retest on the instrument with an interval of 5 days. This testing resulted in a correlation coefficient of 0.38. The nurse researcher interprets this as indicating: instrument is stable; high reliability. instrument is stable; low reliability. instrument is unstable; high reliability. instrument is unstable; low reliability.
ANS: D The low correlation coefficient reveals an unstable instrument and low reliability. The low correlation coefficient reveals an unstable instrument and low reliability. The low correlation coefficient reveals an unstable instrument and low reliability. The low correlation coefficient reveals an unstable instrument and low reliability.
Which type of validity is demonstrated by administering a test in which all items relate to wound care, and then evaluating student performance in caring for patients with wounds in the clinical setting? Face validity Content validity Construct validity Criterion-related validity
ANS: D This does not illustrate face validity, or the instrument appears to be measuring the construct. This does not illustrate content validity, representing the universe of content or domain of a given construct. Construct validity is characterized by the test measuring theoretical construct or trait. This validity type connects the subject's performance with his or her actual behavior.
A newly developed instrument is found to have a Cronbach's alpha of 0.82. The nurse researcher interprets this to mean: the instrument has no internal consistency. the instrument has a low degree of internal consistency. the instrument has a moderate degree of internal consistency. the instrument has a relatively high degree of internal consistency.
ANS: D This is not an accurate interpretation of the example. The example evidences a high degree of internal consistency. The example evidences a high degree of internal consistency. Alphas above 0.70 provide very good or relatively high evidence for supporting the internal consistency of the instrument.
