Reliability and Validity Chong Ho Yu, Ph.D.Conventional views of reliability (AERA et al., 1985) Temporal stability: the same form of a test on two or more separate occasions to the same group of examinees (Test-retest). On many occasions this approach is not practical because repeated measurements are likely to change the examinees. For example, the examinees will adapt the test format and thus tend to score higher in later tests. Hence, careful implementation of the test-retest approach is strongly recommendation (Yu, 2005). Form equivalence: two different forms of test, based on the same content, on one occasion to the same examinees (Alternate form). After alternate forms have been developed, it can be used for different examinees. It is very common in high-stake examination for pre-empting cheating. A examinee who took Form A earlier could not share the test items with another student who might take Form B later, because the two forms have different items. Internal consistency: the coefficient of test scores obtained from a single test or survey (Cronbach Alpha, KR20, Spilt-half). For instance, lets say respondents are asked to rate statements in an attitude survey about computer anxiety. One statement is "I feel very negative about computers in general." Another statement is "I enjoy using computers." People who strongly agree with the first statement should be strongly disagree with the second statement, and vice versa. If the rating of both statements is high or low among several respondents, the responses are said to be inconsistent and patternless. The same principle can be applied to a test. When no pattern is found in the students
responses, probably the test is too difficult and students just guess the answers randomly. Reliability is a necessary but not sufficient condition for validity. For instance, if the needle of the scale is five pounds away from zero, I always over-report my weight by five pounds. Is the measurement consistent? Yes, but it is consistently wrong! Is the measurement valid? No! (But if it under-reports my weight by five pounds, I will consider it a valid measurement)Performance, portfolio, and responsive evaluations, where the tasks varysubstantially from student to student and where multiple tasks may beevaluated simultaneously, are attacked for lacking reliability. One of thedifficulties is that there are more than one source of measurement errors inperformance assessment. For example, the reliability of writing skill test scoreis affected by the raters, the mode of discourse, and several other factors(Parkes, 2000).Replications as unification: Users may be confused by the diversity ofreliability indices. Nevertheless, different types of reliability measures share acommon thread: What constitutes a replication of a measurement procedure?(Brennan, 2001) Take internal consistency as an example. This measure is usedbecause it is convenient to compute the reliability index based upon datacollected from one occasion. However, the ultimate inference should go beyondone single testing occasion to others (Yu, 2005). In other words, anyprocedures for estimating reliability should attempt to mirror a result basedupon full-length replications.
Conventional views of validity (Cronbach, 1971) Face validity: Face validity simply means the validity at face value. As a check on face validity, test/survey items are sent to teachers to obtain suggestions for modification. Because of its vagueness and subjectivity, psychometricians have abandoned this concept for a long time. However, outside the measurement arena, face validity has come back in another form. While discussing the validity of a theory, Lacity and Jansen (1994) defines validity as making common sense, and being persuasive and seeming right to the reader. For Polkinghorne (1988), validity of a theory refers to results that have the appearance of truth or reality. The internal structure of things may not concur with the appearance. Many times professional knowledge is counter-common sense. The criteria of validity in research should go beyond "face," "appearance," and "common sense." Content validity: draw an inference from test scores to a large domain of items similar to those on the test. Content validity is concerned with sample-population representativeness. i.e. the knowledge and skills covered by the test items should be representative to the larger domain of knowledge and skills. For example, computer literacy includes skills in operating system, word processing, spreadsheet, database, graphics, internet, and many others. However, it is difficult, if not impossible, to administer a test covering all aspects of computing. Therefore, only several tasks are sampled from the population of computer skills.
Content validity is usually established by content experts. Take computerliteracy as an example again. A test of computer literacy should be written orreviewed by computer science professors because it is assumed that computerscientists should know what are important in his discipline. By the first glance,this approach looks similar to the validation process of face validity, but yetthere is a difference. In content validity, evidence is obtained by looking foragreement in judgments by judges. In short, face validity can be established byone person but content validity should be checked by a panel.However, this approach has some drawbacks. Usually experts tend to take theirknowledge for granted and forget how little other people know. It is notuncommon that some tests written by content experts are extremely difficult.Second, very often content experts fail to identify the learning objectives of asubject. Take the following question in a philosophy test as an example:What is the time period of the philosopher Epicurus? a. 341-270 BC b. 331-232 BC c. 280-207 BC d. None of the aboveThis type of question tests the ability of memorizing historical facts, but notphilosophizing. The content expert may argue that "historical facts" areimportant for a student to further understand philosophy. Lets change thesubject to computer science and statistics. Look at the following two questions:When was the founder and CEO of Microsoft, William Gates III born? a. 1949 b. 1953
c. 1957 d. None of the aboveWhich of the following statement is true about ANOVA a. It was invented by R. A. Fisher in 1914 b. It was invented by R. A. Fisher in 1920 c. It was invented by Karl Pearson in 1920 d. None of the aboveIt would be hard pressed for any computer scientist or statistician to accept thatthe above questions fulfill content validity. As a matter of fact, thememorization approach is a common practice among instructors.Further, sampling knowledge from a larger domain of knowledge involvessubjective values. For example, a test regarding art history may include manyquestions on oil paintings, but less questions on watercolor paintings andphotography because of the perceived importance of oil paintings in art history.Content validity is sample-oriented rather than sign-oriented. A behavior isviewed as a sample when it is a subgroup of the same kind of behaviors. On theother hand, a behavior is considered a sign when it is an indictor or a proxy of aconstruct. (Goodenough, 1949). Construct validity and criterion validity, whichwill be discussed later, are sign-oriented because both of them indicatebehaviors different from those of thetest.Criterion: draw an inference from testscores to performance. A high score of
a valid test indicates that the tester has met the performance criteria.Regression analysis can be applied to establish criterion validity. Anindependent variable could be used as a predictor variable and a dependentvariable, the criterion variable. The correlation coefficient between them iscalled validity coefficients.For instance, scores of the driving test by simulation is the predictor variablewhile scores of the road test is the criterion variable. It is hypothesized that ifthe tester passes the simulation test, he/she should meet the criterion of being asafe driver. In other words, if the simulation test scores could predict the roadtest scores in a regression model, the simulation test is claimed to have a highdegree of criterion validity.In short, criterion validity is about prediction rather than explanation.Predication is concerned with non-casual or mathematical dependence where asexplanation is pertaining to causal or logical dependence. For example, one canpredict the weather based on the height of mercury inside a thermometer. Thus,the height of mercury could satisfy the criterion validity as a predictor.However, one cannot explain why the weather changes by the change ofmercury height. Because of this limitation of criterion validity, an evaluator hasto conduct construct validation.Construct: draw an inference formtest scores to a psychologicalconstruct. Because it is concerned withabstract and theoretical construct,construct validity is also knownas theoretical construct.According to Hunter and Schmidt(1990), construct validity is a
quantitative question rather than a qualitative distinction such as "valid" or"invalid"; it is a matter of degree. Construct validity can be measured by thecorrelation between the intended independent variable (construct) andthe proxy independent variable (indicator, sign) that is actually used.For example, an evaluator wants to study the relationship between generalcognitive ability and job performance. However, the evaluator may not be ableto administer a cognitive test to every subject. In this case, he can use a proxyvariable such as "amount of education" as an indirect indicator of cognitiveability. After he administered a cognitive test to a portion of all subjects andfound a strong correlation between general cognitive ability and amount ofeducation, the latter can be used to the larger group because its constructvalidity is established.Other authors (e.g. Angoff,1988; Cronbach & Quirk, 1976) argue that constructvalidity cannot be expressed in a single coefficient; there is no mathematicalindex of construct validity. Rather the nature of construct validity is qualitative.There are two types of indictors: o Reflective indictor: the effect of the construct. o Formative indictor: the cause of the construct.When an indictor is expressed in terms of multiple items of aninstrument, factor analysis is used for construct validation.Test bias is a major threat against construct validity, and therefore test biasanalyses should be employed to examine the test items (Osterlind, 1983).The presence of test bias definitely affects the measurement of thepsychological construct. However, the absence of test bias does not guaranteethat the test possesses construct validity. In other words, the absence of test biasis a necessary, but isnt a sufficient condition.
Construct validation as unification: The criterion and the content models tends to be empirical-oriented while the construct model is inclined to be theoretical. Nevertheless, all models of validity requires some form of interpretation: What is the test measuring? Can it measure what it intends to measure? In standard scientific inquiries, it is important to formulate an interpretative (theoretical) framework clearly and then to subject it to empirical challenges. In this sense, theoretical construct validation is considered functioning as a unified framework for validity (Kane, 2001).A modified view of reliability (Moss, 1994) There can be validity without reliability if reliability is defined as consistency among independent measures. Reliability is an aspect of construct validity. As assessment becomes less standardized, distinctions between reliability and validity blur. In many situations such as searching faculty candidate and conferring graduate degree, committee members are not trained to agree on a common set of criteria and standards Inconsistency in students performance across tasks does not invalidate the assessment. Rather it becomes an empirical puzzle to be solved by searching for a more comprehensive interpretation. Initial disagreement (e.g., among students, teachers, and parents in responsive evaluation) would not invalidate the assessment. Rather it would provide an impetus for dialog.Li (2003) argued that the preceding view is incorrect: The definition of reliability should be defined in terms of the classical test theory: the squared correlation between observed and true scores or the proportion of true variance in obtained test scores.
Reliability is a unitless measure and thus it is already model-free or standard- free. It has been a tradition that multiple factors are introduced into a test to improve validity but decrease internal-consistent reliability.An extended view of Mosss reliability (Mislevy, 2004) Being inspired by Moss, Mislevy went further to ask whether there can be reliability without reliability (indices). By blending psychometrics and Hermeneutics, in which a holistic and integrative approach to understand the whole in light of its parts is used, Mislevy demanded psychometricians to think about what they intend to make inferences about. In many cases we dont present just one argument; rather problem solving involves arguments or chains of reasoning with massive evidence. Off-the-shelf inferential machinary (e.g. compute reliability indices) may fail if we quantify things or tasks that we dont know much about. Probability-based reasoning to more complex assessments based upon cognitive psychology is needed.A radical view of reliability (Thompson et al, 2003) Reliability is not a property of the test; rather it is attached to the property of the data. Thus, psychomterics is datammetrics. Tests are not reliable. It is important to explore reliability in virtually all studies. Reliability generalization, which can be used in a meta-analysis application similar to validity generalization, should be implemented to assess variance in measurement error across studies.
An updated perspective of reliability (Cronbach, 2004)In a 2004s article, Lee Cronbach, the inventor of Cronbach Alpha as a way ofmeasuring reliability, reviewed the historical development of Cronbach Alpha. Heasserted, "I no longer regard the formula (of Cronbach Alpha) as the most appropriateway to examine most data. Over the years, my associates and I developed the complexgeneraliability (G) theory" (p. 403). Discussion of the G theory is beyond the scope ofthis document. Nevertheless, Cronbach did not object use of Cronbach Alpha but herecommended that researchers should take the following into consideration whileemploying this approach: Standard error of measurement: It is the most important piece of information to report regarding the instrument, not a coefficient. Independence of sampling Heterogeneity of content How the measurement will be used: Decide whether future uses of the instrument are likely to be exclusively for absolute decisions, for differential decisions, or both. Number of conditions for the testA critical view of validity (Pedhazur & Schmelkin,1991) Content validity is not a type of validity at all because validity refers to inferences made about scores, not to an assessment of the content of an instrument. The very definition of a construct implies a domain of content. There is no sharp distinction between test content and test construct.A modified view of validity (Messick, 1995)
The conventional view (content, criterion, construct) is fragmented and incomplete,especially because it fails to take into account both evidence of the value implicationsof score meaning as a basis for action and the social consequences of score use.Validity is not a property of the test or assessment, but rather of the meaning of thetest scores. Content: evidence of content relevance, representativeness, and technical quality Substantive: theoretical rationale Structural: the fidelity of the scoring structure Generalizability: generalization to the population and across populations External: applications to multitrait-multimethod comparison Consequential: bias, fairness, and justice; the social consequence of the assessment to the societyCritics argued that consequences should not be a component of validity because testdevelopers should not be held responsible for the consequences of misuse;accountability should lie with the misuser. Messick (1998) counter-argued that socialconsequences of score interpretation include the value implications of the constructlabel, which may or may not commensurate with the constructs trait implications andneed to be addressed in appraising score meaning. While test developers should not beaccountable to misuse of tests, they should still pay attention to the unanticipatedconsequences of legitimate score interpretation.A different view of reliability and validity (Salvucci, Walter, Conley, Fink, &Saba (1997)Some scholars argue that the traditional view that "reliability is a necessary but not asufficient condition of validity" is incorrect. This school of thought conceptualizesreliability as invariance and validity asunbiasedness. A sample statistic may have anexpected value over samples equal to the population parameter (unbiasedness), but
have very high variance from a small sample size. Conversely, a sample statistic canhave very low sampling variance but have an expected value far departed from thepopulation parameter (high bias). In this view, a measure can be unreliable (highvariance) but still valid (unbiased).Population parameter (Red line) = Sample Population parameter (Red line) <> Samplestatistic (Yellow line) --> unbiased statistic (Yellow line) --> BiasedHigh variance (Green line) low variance (Green line)Unreliable but valid Invalid but reliableCaution and adviceThere is a common misconception that if someone adopts a validated instrument,he/she does not need to check the reliability and validity with his/her own data.Imagine this: When I buy a drug that has been approved by FDA and my friend asksme whether it heals me, I tell him, "I am taking a drug approved by FDA andtherefore I dont need to know whether it works for me or not!" A responsibleevaluator should still check the instruments reliability and validity with his/her ownsubjects and make any modifications if necessary.Low reliability is less detrimental to the performance pretest. In the pretest wheresubjects are not exposed to the treatment and thus are unfamiliar with the subjectmatter, a low reliability caused by random guessing is expected. One easy way to
overcome this problem is to include "I dont know" in multiple choices. In anexperimental settings where students responses would not affect their final grades, theexperimenter should explicitly instruct students to choose "I dont know" instead ofmaking a guess if they really dont know the answer. Low reliability is a signal of highmeasurement error, which reflects a gap between what students actually know andwhat scores they receive. The choice "I dont know" can help in closing this gap.Last Updated: 2008ReferencesAmerican Educational Research Association, American Psychological Association, &National Council on Measurement in Education. (1985). Standards for educationaland psychological testing. Washington, DC: Authors.Angoff, W. H. (1988). Validity: An evolving concept. In H. Wainer & H. I. Braun(Eds.), Test validity.Hillsdale, NJ: Lawrence Erlbaum.Brennan, R. (2001). An essay on the history and future of reliability from theperspective of replications.Journal of Educational Measurement, 38, 295-317.Cronbach, L. J. (1971). Test validation. In R. L. Thorndike (Ed.). EducationalMeasurement (2nd Ed.). Washington, D. C.: American Council on Education.Cronbach, L. J. (2004). My current thoughts on Coefficient Alpha and successorprocedures. Educational and Psychological Measurement, 64, 391-418.Cronbach, L. J. & Quirk, T. J. (1976). Test validity. In International Encyclopedia ofEducation. New York: McGraw-Hill.Goodenough, F. L. (1949). Mental testing: Its history, principles, andapplications. New York: Rinehart.
Hunter, J. E.; & Schmidt, F. L. (1990). Methods of meta-analysis: Correcting errorand bias in research findings. Newsbury Park: Sage Publications.Kane, M. (2001). Current concerns in validity theory. Journal of educationalMeasurement, 38, 319-342.Lacity, M.; & Jansen, M. A. (1994). Understanding qualitative data: A framework oftext analysis methods.Journal of Management Information System, 11, 137-160.Li, H. (2003). The resolution of some paradoxes related to reliability andvalidity. Journal of Educational and Behavioral Statistics, 28, 89-95.Messick, S. (1995). Validity of psychological assessment: Validation of inferencesfrom persons responses and performance as scientific inquiry into scoringmeaning. American Psychologist, 9, 741-749.Messick, S. (1998). Test validity: A matter of consequence. Social IndicatorsResearch, 45, 35-44.Mislevy, R. (2004). Can there be reliability without reliability? Journal of Educationaland Behavioral Statistics, 29, 241-244.Moss, P. A. (1994). Can there be validity without reliability? Educational Researcher,23, 5-12.Osterlind, S. J. (1983). Test item bias. Newbury Park: Sage Publications.Parkes, J. (2000). The relationship between the reliability and cost of performanceassessments. Education Policy Analysis Archives, 8. [On-line] AvailableURL: http://epaa.asu.edu/epaa/v8n16/Pedhazur, E. J.; & Schmelkin, L. P. (1991). Measurement, design, and analysis: Anintegrated approach.Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers.
Polkinghorne, D. E. (1988). Narrative knowing and the human sciences. Albany: StateUniversity of New York Press.Salvucci, S.; Walter, E., Conley, V; Fink, S; & Saba, M. (1997). Measurement errorstudies at the National Center for Education Statistics. Washington D. C.: U. S.Department of EducationThompson, B. (Ed.) (2003). Score reliability: Contemporary thinking on reliabilityissues. Thousand Oaks: Sage.Yu, C. H. (2005). Test-retest reliability. In K. Kempf-Leonard (Ed.). Encyclopedia ofSocial Measurement, Vol. 3 (pp. 777-784). San Diego, CA: Academic Press.Questions for discussionPick one of the following cases and determine whether the test or the assessment isvalid. Apply the concepts of reliability and validity to the situation. These cases maybe remote to this cultural context. You may use your own example. 1. In ancient China, candidates for government officials had to take the examination regarding literature and moral philosophy, rather than public administration. 2. Before July 1, 1997 when Hong Kong was a British colony, Hong Kong doctors, including specialists, who graduated from non-Common Wealth medical schools had to take a general medical examination covering all general areas in order to be certified. Navigation Index
Simplified Navigation Table of Contents Search Engine Contact