Chapter 10 - Clarifying Measurement and Data Collection in Quantitative Research

An instrument that detects the presence of streptococcal antibodies is negative. Before reporting the absence of disease, the researcher must determine whether this result might be a a. false negative. b. false positive. c. gold standard. d. true positive.

ANS: A A false negative may indicate that a disease is not present when it is. If a test report is negative, the researcher must ensure that the test does not have a high likelihood of false negative results. A false positive means that a disease may not be present when the test result is positive. The gold standard is the most accurate means of currently diagnosing a disease and serves as the basis for comparison for other tests. A true positive means that a positive test result means that the result indicates a positive diagnosis.

A researcher uses a checklist of questions to ask participants while collecting data for a study. This represents which type of data collection? a. Structured interview b. Structured questionnaire c. Unstructured interview d. Unstructured questionnaire

ANS: A A structured interview occurs when the data collector uses a checklist or list of specific questions while interviewing the subject. A structured questionnaire is a tool that the subject completes that uses a checklist or list of specific questions. An unstructured interview occurs when the data collector does not use a checklist or list of specific questions while interviewing the subject. An unstructured questionnaire is a tool that the subject completes that does not use a checklist or list of specific questions.

Which is true about a test that is highly specific? a. It is good at identifying patients who do not have a disease. b. It is good at identifying patients who have a disease. c. It results in an increased number of false negatives. d. It results in an increased number of false positives.

ANS: A A test that is highly sensitive is very good at identifying patients who do not have a disease. A test that is highly sensitive is very good at identifying the presence of disease. Tests that are highly specific have a low percentage of false negatives. Tests that are highly specific have a low percentage of false positives.

The type of validity that focuses on the extent to which an individual's score on an instrument or scale can be used to estimate his or her present or concurrent performance on another variable or criterion is known as a. concurrent validity. b. content validity. c. construct validity. d. predictive validity.

ANS: A Concurrent validity focuses on the extent to which an individual's score on an instrument or scale can be used to estimate his or her present or concurrent performance on another variable or criterion. Content validity measures the extent to which the measurement method includes all the major elements relevant to the construct being measured. Construct validity focuses on determining whether the instrument actually measures the theoretical construct that it purports to measure, which involves examining the fit between the conceptual and operational definitions of a variable. Predictive validity is the extent to which an individual's score on a scale or instrument can be used to predict future performance or behavior on a criterion.

A researcher evaluates whether a tool addresses the major elements of the construct being studied. This helps to ensure a. content validity. b. equivalence. c. internal consistency. d. readability level.

ANS: A Content validity measures the extent to which the measurement method includes all the major elements relevant to the construct being measured. Equivalence and internal consistency are both measures of reliability. Readability focuses on the study participants' ability to read and comprehend the content of an instrument.

A researcher conducts a study to examine the effects of breastfeeding on infant weight at age 6 months. Which type of measurement is used to measure the infants' weight? a. Direct b. Indirect c. Nominal d. Ordinal

ANS: A Direct measurement allows for measures of concrete factors, such as a person's height or weight. Indirect measurement measures abstract ideas or characteristics and often must use more than one measure to measure all aspects of the concept. Nominal level measurement is used when data can be organized into categories of a defined property, such as a disease diagnosis, but when categories cannot be rank ordered. Ordinal level measurement is used when data collected can be rank ordered.

A researcher tests a measurement tool's validity by administering it to a group of subjects who don't have health insurance and a group who have health insurance to examine attitudes about access to health care. The researcher is looking for evidence of validity from a. contrasting groups. b. convergence. c. divergence. d. readability.

ANS: A Evidence of validity from contrasting groups can be tested by identifying groups that are expected or known to have contrasting scores on an instrument. Evidence of validity from convergence is determined when a relatively new instrument is compared with an existing instrument that measures the same construct. Evidence of validity from divergence is determined by asking subjects to complete tools that measure opposite concepts. Readability level focuses on the subjects' ability to read and comprehend the content of an instrument.

A researcher administers two instruments known to measure opposite concepts and performs correlational analyses on the results. The original measure has a negative correlation of -0.6 with the new measure. This indicates a. both measures are valid. b. neither measure is valid. c. only the first measure is valid. d. only the second measure is valid.

ANS: A If a divergent measure is negatively correlated with an opposite scale, the validity for each of the instruments is strengthened. A divergent measure that is negatively correlated with an opposite scale strengthens the validity of both measures and does not negate their validity.

A researcher studying the effects of an intervention on symptoms measures the time from intervention to absence of symptoms and reports this in the number of days. This measure represents which level of measurement? a. Interval b. Nominal c. Ordinal d. Ratio

ANS: A Interval level measurement uses interval scales, which have equal numerical distances between intervals. Nominal level measurement is used when data can be organized into categories of a defined property, such as a disease diagnosis, but when categories cannot be rank ordered. Ordinal level measurement is used when data collected can be rank ordered. Ratio level measurement uses all aspects of other types of measurement: mutually exclusive categories, exhaustive categories, ordered ranks, equally spaced intervals, and a continuum of values.

A researcher may use which strategy to reduce the potential for measurement error when evaluating obesity in study subjects? a. Measure weight, abdominal girth, and BMI on all subjects. b. Train multiple data collectors to take measurements. c. Use scales in several different clinical settings to obtain data. d. Utilize a single measure, such as BMI, applied to all subjects.

ANS: A Obesity is less precise than some concepts, so multiple methods to measure obesity should be used to minimize measurement error. Using multiple data collectors and multiple instruments, such as scales, increases the risk of measurement error. Using multiple data collectors and multiple instruments, such as scales, increases the risk of measurement error. Obesity is less precise than some concepts, so multiple methods to measure obesity should be used to minimize measurement error.

Which statement is true about using a questionnaire to collect data? a. Questions are presented consistently with less chance for bias than with an interview. b. Questionnaires are generally more expensive than interviewing techniques. c. Subjects usually give more in-depth answers when responding to questionnaires. d. The response rate is generally higher than when using interviews to collect data.

ANS: A Questionnaires contain the same questions, so there is increased consistency and less interviewer bias than with interviews, in which the questions may vary and there may be interviewer bias. Questionnaires are less costly to administer than interviews. Subjects tend to give answers that are not as in depth as those they give when being interviewed. Questionnaires generally have a lower response rate than interviews.

A researcher wishes to evaluate the reliability of a visual analog pain scale in a group of postoperative patients. The best way to do this is to compare results a. to results of the test when given to another set of subjects. b. to results from the same subjects using the FACES rating scale. c. to results from a previous measure with a different visual analog scale. d. to results from a group of subjects who have minimal or no pain.

ANS: A Reliability of a visual analog scale is best determined by the test-retest method, so the researcher should administer the test to two sets of subjects and compare the results. Comparing results using two different instruments does not determine reliability of the first measure. Comparing results from a previous measure using another version of the scale is a measure of alternate forms reliability. Changing the population does not allow for a test of the reliability of the measurement tool.

Which is an accurate statement about measurement error? a. Some error is always present in any measurement strategy. b. There is no difference between the true measure and what is actually measured. c. Direct measures, which generally are expected to be highly accurate, are not subject to error. d. With indirect measures, the element being measured can be seen directly.

ANS: A Some error is always present in any measurement strategy. Measurement error is the difference between the true measure and what is actually measured. Direct measures, which generally are expected to be highly accurate, are subject to error. With indirect measures, the element being measured cannot be seen directly.

A researcher reports that a measure is highly sensitive when diagnosing patients. The reviewer understands that this test a. has a low percentage of false negatives. b. has a low percentage of false positives. c. has a high percentage of false negatives. d. has a high percentage of false positives.

ANS: A Tests that are highly sensitive have a low percentage of false negatives. A test that is highly specific has a low percentage of false positives.

Which type of scale is most commonly used in a survey of patient satisfaction? a. Likert scale b. Numeric rating scale c. Rating scale d. Visual analog scale

ANS: A The Likert scale is designed to determine the opinions or attitudes of study subjects and is the most commonly used of the various scales. A numeric rating scale is used to measure the degree of a concept on a continuum, such as pain on a scale from 1 to 10. A rating scale is like a numeric rating scale. A visual analog scale measures the strength, magnitude, or intensity of subjective feelings, sensations, or attitudes along a continuum and may be used to assess patient satisfaction, but it is not the most commonly used type of scale.

Which is true about using data from an existing database? a. It allows larger sampling and complex analyses. b. The burden on participants is higher than when primary data collection is used. c. The researcher can draw upon data that are specific to a study. d. Research reports do not have to describe data collection procedures.

ANS: A Using data from existing databases allows for larger sampling at less cost. Researchers can review data from across situations and time periods, allowing for more complex analyses of data. The burden on participants is lower. The researcher must draw on data from other studies. Data collection procedures must be described with any data used.

Which statement is true about validity? a. Validity evaluates an instrument for a specific group or purpose. b. Validity occurs when an instrument is valid across situations. c. Validity measures are absolute. d. Validity measures are independent of reliability.

ANS: A Validity will vary from one sample to another and one situation to another, so validity testing evaluates the use of an instrument for a specific group or purpose. Validity measures are not all-or-nothing but occur on a continuum. Without reliability, a tool cannot be valid.

Which of the following are accurate statements about the data collection process when conducting research? Select all that apply. a. The actual steps of collecting data are specific to each study and depend on the research design, sample, and measurement techniques. b. Study participants may be recruited only at the initiation of data collection, but not throughout the data collection period. c. It is not necessary for the researchers to specify the number and characteristics of subjects who decline to participate in the study. d. The key to accurate data collection in any study is consistency. e. Researchers build controls into their study plan to maximize the influence of intervening forces on the findings.

ANS: A, D The actual steps of collecting data are specific to each study and depend on the research design, sample, and measurement techniques. The key to accurate data collection in any study is consistency. Consistency involves maintaining the data collection pattern for each collection event as it was developed in the research plan. Study participants may be recruited only at the initiation of data collection or throughout the data collection period. Researchers should also specify the number and characteristics of subjects who decline to participate in the study. Researchers build controls into their study plan to minimize the influence of intervening forces on the findings.

Which of the following are true statements about a Likert scale? Select all that apply. a. It is designed to determine the opinions or attitudes of study subjects. b. It is the least commonly used of the scaling techniques. c. The original version of the scale included three response categories. d. Sometimes seven options are given on a response scale, sometimes only four. e. It usually consists of 10 to 20 items, each addressing an element of the concept being measures.

ANS: A, D, E The Likert scale is designed to determine the opinions or attitudes of study subjects. This scale contains a number of declarative statements, with a scale after each statement. Sometimes seven options are given on a response scale, sometimes only four. A Likert scale usually consists of 10 to 20 items, each addressing an element of the concept being measures. The Likert scale is the most commonly used of the scaling techniques. The original version of the scale included five response categories.

What type of information is typically found in an administrative database? a. Data collected for a specific research study b. Demographic information about a population c. Journal accounts of subjects' responses to surveys d. Statistical results of experimental studies

ANS: B Administrative data are collected within clinical agencies and government entities and include demographic, organizational, clinical, and geographical information. Data collected for a specific study are primary data. Journal accounts and statistical results of experimental studies may be primary or secondary data. Journal accounts and statistical results of experimental studies may be primary or secondary data.

Which can be measured using direct measurement? a. Anxiety level b. Blood pressure c. Diagnosis d. Feelings

ANS: B Direct measurement allows for measures of concrete factors, such as a person's height, weight, or blood pressure. Indirect measurement involves measuring abstract concepts such as anxiety or feelings. Nominal level measurement is used when data can be organized into categories of a defined property, such as a disease diagnosis, but when categories cannot be rank ordered. Indirect measurement involves measuring abstract concepts such as anxiety or feelings.

A researcher develops a new instrument to measure coping skills and conducts a pilot study to compare the new tool with an existing tool that measures this concept. This is an example of evidence of validity from a. contrasting groups. b. convergence. c. divergence. d. readability.

ANS: B Evidence of validity from convergence is determined when a relatively new instrument is compared with an existing instrument that measures the same construct. Evidence of validity from contrasting groups can be tested by identifying groups that are expected or known to have contrasting scores on an instrument. Evidence of validity from divergence is determined by asking subjects to complete tools that measure opposite concepts. Readability level focuses on the subjects' ability to read and comprehend the content of an instrument.

A researcher evaluates scores of a paper and pencil instrument with dichotomous variables using the Kuder-Richardson formula. This is a test of a. alternate forms reliability. b. internal consistency reliability. c. interrater reliability. d. test-retest reliability.

ANS: B Internal consistency reliability addresses the correlation of each question to other questions within the scale. Alternate forms reliability seeks to determine whether two forms of an instrument measure the same attributes. Interrater reliability measures whether two or more observers measure the same attributes. Test-retest reliability measures whether the same attribute is consistent from one testing time to another.

A researcher conducts a study to identify the relationship of lifestyle choices to the development of chronic diseases. The researcher surveys subjects and identifies diabetes mellitus, coronary artery disease, and renal disease in study subjects. These measures represent which level of measurement? a. Interval b. Nominal c. Ordinal d. Ratio

ANS: B Nominal level measurement is used when data can be organized into categories of a defined property, such as a disease diagnosis, but when categories cannot be rank ordered. Interval level measurement uses interval scales, which have equal numerical distances between intervals. Ordinal level measurement is used when data collected can be rank ordered. Ratio level measurement uses all aspects of other types of measurement: mutually exclusive categories, exhaustive categories, ordered ranks, equally spaced intervals, and a continuum of values.

Which generally is an advantage of using primary data in a research study? a. It allows researchers to conduct more longitudinal studies. b. It has greater specificity to the study purpose. c. It puts less burden on study participants. d. It reduces the cost of data collection.

ANS: B Primary data are collected for a study and are specific to that study. Secondary data are collected by other researchers and then stored and can contribute to longitudinal studies. Primary data collection puts more of a burden on study participants. Primary data collection is more costly than secondary data collection.

Which data represent the lowest form of measurement? a. Levels of anxiety on a scale from mild to severe b. Race and ethnicity classification c. Temperature of subjects after an intervention d. Weight, height, and body mass index

ANS: B Race and ethnicity represent nominal data, which are considered the lowest form of measurement. Levels of anxiety represent ordinal data. Temperature represents interval data. Weight, height, and BMI represent ratio data.

A researcher notes a reliability score of 0.75 between two observers in a clinical drug trial evaluating whether the drug decreases anxiety. This score means that a. there is 25% random error. b. there is 75% random error. c. this is an acceptable measure of reliability. d. this measurement tool is not reliable.

ANS: B The interrater reliability value represents reliability. A 0.75 measure indicates 75% reliability and 25% random error. Generally, interrater reliability is best at 0.90, and any value less than 0.80 should generate serious concern about the reliability of the data or the data gatherer. This is a measure of interrater reliability and may not necessarily reflect on the reliability of the tool.

In a study involving multiple data collectors, the researcher helps to ensure consistency in data collection by a. describing the study participant recruitment process in the research summary. b. detailing the training process of the data collectors in the research report. c. discarding data that includes even minor deviations in data collection methods. d. minimizing the influence of intervening forces on the study findings.

ANS: B To address consistency, the researcher must detail how data collectors are trained and how interrater reliability is achieved in the study report. Describing the recruitment process is important but does not contribute to consistency in data collection. If deviations in data collection methods occur, they should be discussed, but the results should not be discarded. Minimizing the influence of intervening forces helps to control the study design.

Which is true about a test that is highly sensitive for detecting a disease? a. There is a high probability of having a false negative result. b. There is a high probability of having a false positive result. c. There is a low probability of having a false negative result. d. There is a low probability of having a false positive result.

ANS: C A test that is highly sensitive is very good at identifying the presence of disease and has a low percentage of false negatives. Tests that are highly specific have a low percentage of false positives. Tests that are highly sensitive have a low percentage of false positives. Tests that are highly specific have a low percentage of false positives.

A screening test has a specificity of 92%. This means that there is a(n) a. 92% chance of a true positive. b. 92% chance of a false positive. c. 8% chance of a false positive. d. 8% chance of a true positive.

ANS: C A test with a specificity rating of 92% is very good at identifying patients who do not have a disease. If the test is very specific, there is a low chance of false positives. In this case the chance of a false positive is 8%. A 92% specificity rating indicates a 92% chance that the test is negative and the patient does not have the disease.

A nursing professor administers three versions of a final examination to a class of students and compares the examinations for reliability. This is a measure of a. homogeneity. b. interrater reliability. c. equivalence. d. test-retest reliability.

ANS: C Alternate forms reliability, or stability, seeks to determine whether two forms of an instrument measure the same attributes. Homogeneity, or internal consistency reliability, addresses the correlation of each question to other questions within the scale. Interrater reliability measures whether two or more observers measure the same attributes. Test-retest reliability measures whether the same attribute is consistent from one testing time to another.

A researcher administers a scale measuring self-confidence and one measuring self-doubt to a group of subjects and performs a correlational analysis of the results. This is a test of validity from a. contrasting groups. b. convergence. c. divergence. d. readability.

ANS: C Evidence of validity from divergence is determined by asking subjects to complete tools that measure opposite concepts. Evidence of validity from contrasting groups can be tested by identifying groups that are expected or known to have contrasting scores on an instrument. Evidence of validity from convergence is determined when a relatively new instrument is compared with an existing instrument that measures the same construct. Readability level focuses on the subjects' ability to read and comprehend the content of an instrument.

A researcher studying children identifies the following age groups: Birth to age 12 months 12 months to 3 years 3 years to 5 years 5 years to 12 years 12 years to 18 years These categories are not a. empirical. b. equal. c. exclusive. d. exhaustive.

ANS: C Exclusive means a datum must fit into only one of the categories. In this example, someone age 5 can fit into two categories. Empirical is a term implying some objective reality. The nonequal divisions would only relate to the level of measurement and not pose a problem. The example gives age ranges that exhaust all possibilities for ages of children.

In a study measuring oxygen saturation levels in infants, the researcher notes that data collectors record low saturation levels that occur when the heart rate is erratic and the infants are irritable and actively moving around. This will likely result in which type of measurement error? a. Environmental b. Equipment c. Interpretation d. User

ANS: C Interpretation error occurs when signals transmitted from the equipment can result in misinterpretation, such as low saturation levels that occur when the equipment is not getting an accurate reading. User errors are caused by the person using the equipment and can include variations by the same user, different users, or changes in supplies used to operate the equipment. Environmental error includes things such as temperature, barometric pressure, or static electricity that might affect how well the equipment works or how the subject responds. Equipment error may be related to calibration or the stability of the equipment.

In a pilot study to help train a group of data collectors who will evaluate children's pain using a FACES rating scale, the researcher examines the reliability of the scores of each data collector. This is an example of which type of reliability measure? a. Alternate forms reliability b. Internal consistency reliability c. Interrater reliability d. Test-retest reliability

ANS: C Interrater reliability measures whether two or more observers measure the same attributes. Alternate forms reliability seeks to determine whether two forms of an instrument measure the same attributes. Internal consistency reliability addresses the correlation of each question to other questions within the scale. Test-retest reliability measures whether the same attribute is consistent from one testing time to another.

Among subjects with coronary artery disease, a researcher identifies three levels of symptoms which can be categorized by severity. This represents which level of measurement? a. Interval b. Nominal c. Ordinal d. Ratio

ANS: C Ordinal level measurement is used when data collected can be rank ordered. Interval level measurement uses interval scales, which have equal numerical distances between intervals. Nominal level measurement is used when data can be organized into categories of a defined property, such as a disease diagnosis, but when categories cannot be rank ordered. Ratio level measurement uses all aspects of other types of measurement: mutually exclusive categories, exhaustive categories, ordered ranks, equally spaced intervals, and a continuum of values.

A researcher evaluating the weight of subjects who are obese ensures that the scales used are maintained and calibrated according to the manufacturer's recommendations. This is to ensure a. accuracy. b. convergence. c. precision. d. validity.

ANS: C Precision is the degree of consistency or reproducibility of measurements made with physiological instruments and is comparable to reliability in psychosocial measures. Accuracy is comparable to validity, addressing the extent to which the instrument measures what it is supposed to measure. Evidence of validity from convergence is determined when a relatively new instrument is compared with an existing instrument that measures the same construct. Validity addresses the extent to which the instrument measures what it is supposed to measure.

A researcher performs a series of pilot studies to evaluate whether a measurement tool produces consistent results. This is an evaluation of a. accuracy. b. precision. c. reliability. d. validity.

ANS: C Reliability is concerned with the consistency of a measurement method. Accuracy and precision are part of the evaluation of physiological measurement tools. Accuracy and precision are part of the evaluation of physiological measurement tools. Validity is concerned with how well the instrument reflects the abstract concept being examined.

Which is a true statement about measurement theory? a. A measurement method used by one individual will consistently produce different results when used by another individual. b. Measurement theory was recently developed by musicians, statisticians, and other scholars to guide how things are measured. c. The purpose of measurement is to obtain trustworthy data that can be used to address the study purpose and objectives, questions, or hypotheses. d. The levels of measurement, from low to high, are ratio, interval, nominal, and ordinal.

ANS: C The purpose of measurement is to obtain trustworthy data that can be used to address the study purpose and objectives, questions, or hypotheses. The rules of measurement promote consistency in how individuals perform measurements; so a measurement method used by one individual will consistently produce similar results when used by another individual. Measurement theory was developed many years ago, by mathematicians, statisticians, and other scholars to guide how things are measured. The levels of measurement, from low to high, are nominal, ordinal, interval, and ratio.

The nurse researcher understands that the levels of measurement, from low to high, are a. nominal, interval, ratio, and ordinal. b. nominal, ordinal, ratio, and interval. c. nominal, ordinal, interval, and ratio. d. nominal, interval, ordinal, and ratio.

ANS: C The traditional levels of measurement were developed by Stevens (1946), who organized the rules for assigning numbers to objects so that a hierarchy in measurement was established. The levels of measurement from low to high are nominal, ordinal, interval, and ratio (NOIR).

Which possible outcome of a screening test for a disease indicates accurately that a disease is not present? a. True positive b. False positive c. True negative d. False negative

ANS: C True negative indicates accurately that a disease is not present. True positive is an accurate identification of the presence of a disease. False positive indicates that a disease is present when it is not. False negative indicates that a disease is not present when it is.

Which is important when using observational measurement to produce reliable data? a. Data collectors will record observations in their own words. b. Each subject should have different observers to record responses. c. Observations will be spontaneously recorded as they occur. d. Researchers will develop checklists of notable behaviors for observers to use.

ANS: D Checklists of behaviors help to indicate whether specific behaviors occur and help to structure observations. Data collectors must be trained to observe specific behaviors and in how to describe and code those behaviors. Increasing the number of data collectors increases the risk for error. Unstructured observation allows for spontaneous recording of observations but is less reliable than structured observation.

A researcher wishes to study the effects of preoperative teaching on anxiety levels among toddlers and preschoolers undergoing surgery. To reduce the potential for measurement error, the researcher will a. ask study subjects to describe feelings of anxiety. b. develop a new tool to measure anxiety levels in children. c. rely on multiple observers to collect data for this study. d. use a standardized checklist of anxiety behaviors.

ANS: D Measurement will be more precise if researchers use a well-developed, reliable scale. Using a checklist of anxiety behaviors reduces the likelihood of observer error. The technique of asking subjects to describe feelings provides in-depth data, but not precise measures, and is used in qualitative research. A new tool that has not been evaluated for reliability and validity has an increased chance of error. Increasing the number of observers increases the likelihood of error.

A study measuring the effects of a new diuretic medication records hourly urine output of subjects. This measure represents which level of measurement? a. Interval b. Nominal c. Ordinal d. Ratio

ANS: D Ratio level measurement uses all aspects of other types of measurement: mutually exclusive categories, exhaustive categories, ordered ranks, equally spaced intervals, and a continuum of values. Interval level measurement uses interval scales, which have equal numerical distances between intervals. Nominal level measurement is used when data can be organized into categories of a defined property, such as a disease diagnosis, but when categories cannot be rank ordered. Ordinal level measurement is used when data collected can be rank ordered.

A researcher reviews study data about head circumference in newborns and notes that study personnel are measuring from the end of the measuring tape and not from the zero point, which is 1 cm from the end. This is an example of which type of measurement error? a. Indirect b. Random c. Reliability d. Systematic

ANS: D Systematic error is the variation in measurement values from the calculated average that occurs systematically either because the instrument is also measuring something else, or because the instrument is set to a scale that is off the true measure. In this case, the personnel are inadvertently adding a centimeter to the measurements. Indirect error is an error that occurs when measuring indirect concepts. Random error occurs randomly, when the difference between the measured value and the true value is without pattern or direction. Reliability error occurs when a measurement method is not consistent.

A researcher evaluates a measurement tool by administering the tool to a group of subjects every 2 weeks for 8 weeks. This is a measure of the tool's a. alternate forms reliability. b. internal consistency reliability. c. interrater reliability. d. test-retest reliability.

ANS: D Test-retest reliability measures whether the same attribute is consistent from one testing time to another. Alternate forms reliability seeks to determine whether two forms of an instrument measure the same attributes. Internal consistency reliability addresses the correlation of each question to other questions within the scale. Interrater reliability measures whether two or more observers measure the same attributes.

In a study evaluating pulse oximetry measures in infants, a researcher notes that the pulse oximeter probes are attached to infants' feet in different ways. This represents which type of measurement error? a. Environmental b. Equipment c. Subject d. User

ANS: D User errors are caused by the person using the equipment and can include variations by the same user, different users, or changes in supplies used to operate the equipment. Environmental error includes things such as temperature, barometric pressure, or static electricity that might affect how well the equipment works or how the subject responds. Equipment error may be related to calibration or the stability of the equipment. Subject error occurs if the subject alters the equipment or the equipment alters the subject.

When documenting reports that previous research has demonstrated the accuracy of a physiological measure, the researcher is addressing the measure's a. equivalence. b. homogeneity. c. reliability. d. validity.

ANS: D Validity refers to the fact that the instrument measures what it says it will measure. Accuracy is comparable to validity, addressing the extent to which the instrument measures what it is supposed to measure. Equivalence focuses on consistency between two or more observers measuring the same event. Homogeneity examines the extent to which all items in an instrument consistently measure the construct of interest. Reliability is concerned with the consistency of a measurement method.