36 the likert scale revisited

Upload: afvillodres7471

Post on 03-Jun-2018

229 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/12/2019 36 the Likert Scale Revisited

    1/18

    The Likert scale revisited: An alternate versionGerald Albaum

    Market Research Society. Journal of the Market Research Society; Apr 1997; 39, 2; ABI/INFORM Global pg. 331

    A I b a u m T h e ~ i k e r t s e a 1 r v 1 S 1 t e d a n a I t e r n a t e ve r < i o n

    he Likert scale revisitedan ltern te version

    Gerald AlbaumUniversity of Oregon

    Gerald lbawn

    This s tudy examined the effect o f al ternat ive scaleformats on repor t ing o f intensity o f at t i tudes onLikert scales o f agreement . A s tandard one-stageformat and an al ternate two-stage format weretes ted in three separa te studies on samples o funiversity students in th ree countr ies . n general,the two-stage format generated the greatestpercentage o f extreme-posi t ion (Le. most intense)

    responses across scales. A tes t o f predict ive abili tyshowed tha t the two-stage format was a bet terpredic tor o f product preferences. Underlying da tas t ruc tures d id not differ much between the two.

    The Likert scale revisited: an al ternate versionMarketing researchers have available for their use in measuring attitudes orcomponents of attitudes a number of different types of scales or scale formats,including the semantic differential, Stapel scale, Likert scale, Thurstonedifferential scale, and direct rating scales. One of the most widely used formats isthe Likert scale.

    The purpose of this article is to examine the Likert scale in terms of its abilityto detect the intensity of feeling that respondents have about their attitudes. Analternative format s proposed and tested against the standard approach.

    Nature o f the Liker t scaleAttitude is a complex thing. Recent theoretical models of attitudes suggest thatthere are two dimensions - direction (e.g. a positive or negative predispositiontoward an object) and strength (Petty Krosnick in press; Raden 1985). For

    example, a person may like or dislike a product and may hold this attitude with avarying degree of strength or intensity - interpreted as either confidence,certainty, accessibility or conviction (Berger Alwin 1996). Abelson (1988)distinguishes between firmly held attitudes and those that are more superficial byutilising the idea of attitude conviction Berger Alwin (1996) propose thatconviction s a subjective component of an attitude which gauges the extent towhich the attitude is owned or firmly held and reflects the functions the attitude

    331

  • 8/12/2019 36 the Likert Scale Revisited

    2/18

    Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

    J ~ u r n a lI f h e .rV1.arket R e s e a r c h S o c i e t y Vo l u m e 3 4 N u m b e r 2 A p r 1 ) 1 9 9 7

    serves. Our concern in this paper is with the extent to which an attitude is firmlyheld. Consequently, we equate conviction with strength and intensity.

    When a Likert scale is used to measure attitude, its usual or standard formatconsists of a series of statements to which a respondent is to indicate a degree ofagreement or disagreement using the following options: strongly agree, agree,neither agree nor disagree, disagree, strongly disagree. As such the scale purportsto measure direction (by 'agree/disagree') and intensity (by strongly or not) ofattitude. The scale, per se was intended as a summated scale which was thenassumed to have interval scale properties (Likert 1932). This 1evel-ofmeasurement characteristic together with ease of administration and responseexplains its popularity in marketing research applications. The individual scale isnot assumed to be intervally measured although it usually is treated as such. In

    practice, the scale is often used by researchers in marketing as individual scaleitems or as a summated scale based on a small number of scale items (as few astwo or three).

    According to Tourangeau (1984; 1992) with some attitude questions aperson must compute an evaluative judgement whereas with others such ajudgement is simply retrieved. The form of the attitude question may alsoinfluence whether a respondent employs one process or the other. A Likert-typeitem, because it requires a person to rate extent of agreement, may encouragethe retrieval and integration of more detailed information from memory than do

    items calling for a simple evaluation. In a real sense we can view an opinion as averbal expression of an attitude which means that opinions are the means wehave for measuring attitudes.

    The standard Likert scale tends to confound the direction and intensitydimensions of attitude so there may be an under-reporting of the most intenseagreement or disagreement (i.e. the extreme position of the scale). When lookingat extreme positions, we are dealing with the furthermost categories of theresponse set alternatives and not extremeness in amount of direction. In theLikert scheme the furthermost categories represent the most intensity ofresponse. In an analysis-of-variance context the standard Likert scale measuresdirectly the interaction and indirectly, the main effects of direction and intensity.In short, main effects are inferred from the interaction measure.

    An alternative way of presenting the Likert scale is as a two-stage scale(Mager Kluge 1987). The first stage asks whether there is agreement ordisagreement with the statement. The second stage then asks how strongly theperson fe els about the answer provided in the first stage. In a more general sense,the split question technique has been called unfolding and has been used intelephone surveys as an alternative to the use of show cards in face-to-faceinterviews (Groves 1979; Groves Kahn 1979; Miller 1984; Sykes Collins1988). The end result of this split question process for the Likert scale is that thedirection dimension of attitude is being addressed by the first stage and theintensity (or strength) dimension is measured directly by the second stage.Interaction effects can then be measured. This process comes closer to the usualANOVA approach. In addition, Berger Alwitt (1996) present evidence that

    332

  • 8/12/2019 36 the Likert Scale Revisited

    3/18

    Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

    A l b a u m T h e L i k e r [ s c a l e r e v i s i t e d a n < l l t e r n a t c v c r s i o n

    Figure 1Fonna t s for the Likert-type scale

    1 One-stage the traditional format)

    Please indicate your degree of agreement with each of the following statements by circling one ofthe symbols beneath the statement where

    SA=strongly agreeA=agreeN=neither agree nor disagree

    A product ' s price will usually reflect its level of quality.'SA N D SD NO

    II . Two-Stage

    D=disagree

    SD=strongly disagreeNO=no opinion

    For each of the statements listed below indicate first the extent of your agreement and secondhow strongly you feel about your agreement.

    A product s price will usually reflect its level of quality. 'neither agree

    gree ~ r djsagree isagree ~ o o p n o nHow strongly do you feel about your response?

    ery strong ~ t very strong

    attitude strength and direction are unique separable dimensions of an attitude. tis this two-stage approach that is examined by the study reported in this paper(see figure 1).

    What is known about the methodological characteristics of the Likert scale?*Jacoby Matell (1971) examined the issue of number of response alternatives tobe provided. Since there was no effect on the use of categories these researchersconcluded that as little as two or three categories could be used. Such scales aresufficient to meet criteria of test-retest reliability, concurrent validity, andpredictive validity Oacoby Matell 1971). A major result of reducing thecategories in this manner is that only direction s provided. Matell Jacoby(1972) raised a broader question about optimal number of categories whileHulbert Lehmann (1972) argued that the goals or objectives underlying use ofscales should dictate the number of scale categories. A related issue is that ofwhat response categories should be used. Spector (1976) was concerned withthe interval-level measurement issue and attempted to derive scale values foralternative response categories. In contrast, Wyatt Meyers (1987) examinedfour different \vays to label five-point response scales. They found no differencein mean values but the scales did differ on variability. A similar finding emergedfrom a study by Dixon t at 1984).

    Another aspect studied was that of the neutral point (Komorita 1963; GuyNorvell 1977; Garland 1991). There could be a sensitising effect if a respondentexpects to find a neutral position and does not find one. Since composite scorestend not to be affected by inclusion, it has been argued that a neutral position* Our concern s with specific characteristic s of the Likert scalc. Thcre s a body of research that hascompared the Likert scale with other types of scales such as Semantic scales (Ofir e l a i I987) , directranking scales (Conrath, Montazemi Higgins 1987), Thurstone scales (Flamer 1983) and theGuttman scale (Byrne, 1987).

    333

  • 8/12/2019 36 the Likert Scale Revisited

    4/18

    Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

    J ) U r Il a 1 ) f t h e \1 a r k e t e s e a r c h S o c i c t V Vo l II m c 3 9 K u m b e r p r i l l 9 9 7

    should always be provided. In another study, Armstrong (1987) compareddifferent ways to express the midpoint - undecided or neutral. Analysis showeddifferences were negligible and little if any erosion of score appeared to result. A

    neutral position s one that is neither agree nor disagree, and does not includethe categories 'don't know' and 'no opinion.'

    As with any scale and scale format there s concern for systematic errors, onetype of which is form-related (Bardo, Yeager Klingsporn 1982; BardoYeager 1982; Phelps et al 986; Greenleaf 1992b). Form-related errors concernpsychological orientation towards responding to different item formats andinclude the following types:

    Leniency: tendency to rate something too high or too low (i.e. rate in anextreme way). Central tendency: reluctance to give extreme scores. Proxim ity: give similar responses to items that occur close to one another.

    The present study will allow for an appraisal of whether the form-relatedpotential errors might be 'generic' for the Likert scale and Likert-type data.

    Directly related to the present study is the issue of direction and extremeness.Indeed, a question can be raised as to whether there s a difference betweenextremeness and intensity in using a Likert scale. Converse Presser (1986)have argued that the Likert scale confounds extremity (a dimension ofattitudinal position) with intensity (how strongly a position is felt). Althoughintensity and extremity may frequently covary, a person may hold an extremeposition with little feeling, or have a middle of the road position withconsiderable passion; without separate questions for positions and intensity it sdifficult, if ever possible, to separate these dimensions. Peabody (1962) hasargued that there are both theoretical and practical reasons for considering thecomponents of direction and extremeness (i.e. intensity) separately. Theirresearch did not deal specifically with the Likert scale. Similarly, CompeauFranke (1990) explored direction and intensity in the broader category ratingscale. Both sets of researchers agree that composite scores (e.g. that from aLikert scale) do not reflect the intensity dimension well, and separating thedirection and intensity components may allow the researcher to explain morevariance than the composite score where the two dimensions are combined.Furthermore, separating the components may assist the researcher ininterpreting how the respondents used the measure (Compeau Franke 1990).Applying the technique to dual scaling - an approach to correspondence analysis- has been suggested as a way to separate intensity and direction (CompeauFranke 1990; Franke 1985). The present paper presents another approach.* tmust be remembered that in the Likert Scale format extremity and positions onthe scale measure intensity of feeling about direction and not amount ofdirection. Amount of direction in attitude has to be captured by the statementitself.* On a more generallcvcl, Greenleaf (l992a) proposes a method for creating, validating and scoringextreme re ponse style measures,

    334

  • 8/12/2019 36 the Likert Scale Revisited

    5/18

    Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

    A l b a u m : T h e L i k e r t ~ c l c r e v I s I t e d : a n l t e rn te v c r S I o n

    More recent research dealing specifically with the Likert scale and Likerttype data involves adaptive survey designs based on a graded-response LatentTrait Theory model Singh, Howell Rhoads 1990). Operating like an expertsystem, an adaptive survey design d pts the questions asked to each respondentbased upon responses to previous questions. In short, an adaptive survey asksdifferent questions of different respondents, but s able to estimate allrespondents' locations, along the same attitude continuum. Because Likert-typedata provide much information, a smaller number of items may be used for adesign based on the graded-response model than for other survey designs usingthe scale. These researchers suggest efficiencies of the magnitude of 33compared with conventional survey designs Singh, Howell Rhoads 1990,p. 320). Finally, Russell Bobko (1992) examined how characteristics of a

    Likert-type scale affected the power of moderated regression analysis. Theresponse scale was not a true Likert-scale as the categories were of motivationtowards some future behaviour not categories of agreement with statements.

    As mentioned earlier, the present study looks primarily at the issue ofconfounding of attitude components and the reporting of intensity (or extremepositions available) by the scale. It is hypothesised that the standard one-stageLikert scale tends to under-report the most intense position (thus, enhancingcentral leniency errors), compared with a two-stage format. The specifichypothesis was suggested by the results of a small exploratory study of these

    alternative scale formats for the Likert scale (Albaum Murphy 1988). In mostcases, a two-stage approach generated a greater percentage of most intenseposition values than did standard and modified one-stage versions.

    ethod

    Three distinct studies were conducted to generate the data needed to test theoverall hypothesis. Each study was set up as a simple experiment using acompletely randomised design. Two treatment levels were used as shownpreviously in figure 1. Data were collected in different countries at different timeperiods over several months.

    The topic of Study 1 was economic systems. Separate convenience samplesof business students were drawn from two different universities in the UnitedStates (N 1 =64, N n =97), one university in New Zealand (N=121), and oneuniversity in Denmark (N=98). The sample in Denmark consisted of part-timestudents, each of whom s employed full time. This study was done in threecountries to explore whether each scale type tends to be culture-bound (an emic)or culture-free (an etic). Independent of the absolute values that emerge, theLikert scale will possess etic properties as a scale type if results of one-stage andtwo-stage comparison are similar within each country.

    The Likert-scale instrument for the United States sample includedstatements about capitalism and different economic systems. For both theDenmark and New Zealand samples, two additional statements were addedgiving a total of 13 statements. These two statements were added because oftheinterest in them by associates in the countries who assisted in data collection.

    335

  • 8/12/2019 36 the Likert Scale Revisited

    6/18

    Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

    J o u r r . al o f t h e i v i a r k e t R e s e a r c h S o c i e t y Vo l u m e 1 9 N u m b e r A p r i l 9 9 7

    One statement dealt with communism and resource allocation efficiency; theother statement was about size of firm and private enterprise as capitalism. Forthe most part, the final measurement instruments were the same in eachcountry. The instrument used in Denmark was presented to respondents inDanish after two rounds of translation and back translation.

    To remove doubt that might arise about a topic/scale-format interactionStudy 2 was designed around a topic that is quite different from that used inStudy 1 The overall topic of Study 2 was alienation. A convenience sample of176 students in New Zealand responded to a set of 37 statements aboutconsumer alienation from the marketplace (Allison 1978). As in Study 1, acompletely randomised experimental design with two treatment levels (onestage and two-stage Likert scales) was used in Study 2.

    The results of the first two studies were such that a third study was needed toassess which of the formats was best . Study 3 was designed around the topic ofconsumer ethnocentrism. A sample of 50 students in the United Statesresponded to the 17 statements that comprise the CETSCALE (ShimpSharma 1987). This scale is designed to measure consumers ethnocentrictendencies related to purchasing foreign- versus American-made products.Although originally developed and validated with samples of US consumersonly, other research has shown that the CETSCALE is a reliable measure acrossother countries and affords some evidence of validity as well (Netemeyer,

    Durvasula Lichtenstein 1991). In addition to the CETSCALE items, aquestion was included asking for preferences regarding purchasing nineconsumer durable products ranging from electronics to athletic shoes. A ninecategory scale was used, anchored with American-made and foreign-made. Asin the previous two studies, a completely randomised experimental design withtwo treatment levels was used in Study 3.

    esults

    For all three studies responses to the one-stage and two-stage Likert scales were

    compared at the level of the individual scale. Although the Likert Scale issupposed to be used as a summated scale, it often is not used that way byacademic and practitioner marketing researchers. Individual scale item scoresare used for analysis, as are average scale values for the scales comprising aconstruct. In addition, many reported summated scores consist of as few as twoor three scale items. As we pointed out earlier there is evidence in the literaturethat composite scores do not reflect the intensity dimension well. On a morepract ical level, there are times when researchers are interested in groups - suchas market segments - having intense beliefs about an object of concern. Often,this is best picked up at the individual scale level. The two-stage version wasscored t be comparable with the one-stage version. That is, a respondent whoanswered the first stage with agree or disagree and the second stage with notvery strong was scored the same as one who indicated agree or disagree to theone-stage version. Similarly, a respondent who answered the second stage withvery strong was scored the same as one who answered strongly agree or

    336

  • 8/12/2019 36 the Likert Scale Revisited

    7/18

    R e pr o

    d u c e d wi t h

    p er mi s

    s i on

    of

    t h e c o p y r i gh

    t own

    er .F

    ur t h er r e

    pr o

    d u c t i on

    pr oh i b i t e

    d wi t h

    o u t p er mi s

    s i on.

    Table Extreme responses to economic systems, attitudeUnited States samples*Percent of respondents)

    US l l

    Strongly Stronglydisagree agree

    One Two One- Two-AltilUde slatement slage slage'> stage Slage b

    The United States should movetowards socialism 43.0 54.7 .10 2.2 6.7 .16The private enterprise system is best for smallbusiness firms rather than big business finns 10.6 26.3 .01 2.1 18.4 .01The private enterprise economic system issynonymous with small business 4.3 27.0 .01 3.2 12.2 .02In order to achieve economic growthenvironmental laws should be relaxed 38.1 79.5 .01 0.0 3.6 .07Capitalism does not encourage small business 31.9 67.5 .01 2.1 5.0 .30Big business operates best within a frameworkof capitalism rather than private enterprise 6.0 9.9 .31 2 .4 31.0 .01The economic system of capitalism issynonymous with big business 2.2 6.4 .15 2 .2 34.6 .01 Capitalism is socially equitable 9.7 36.0 .01 1.1 12.0 .01Socialism leads to high living standards 28.7 54.5 .01 0.0 5.2 .03

    Capitalism and private enterprise aredifferent economic systems 8.8 30 .3 .01 0.0 7.9 .01Government should guarantee agood standard of living 8 .3 40.2 .01 5 .2 30.5 .01

    *Excludes 'don' t know', and 'no opinion'. aN =49 bN =48 eN =32

    v:>J

  • 8/12/2019 36 the Likert Scale Revisited

    8/18

    R e pr o

    d u c e d wi t h

    p er mi s

    s i on

    of

    t h e c o p y r i gh

    t own

    er .F

    ur t h er r e

    pr o

    d u c t i on

    pr oh i b i t e

    d wi t h

    o u t p er mi s

    s i on.

    \ N\ N Table Extrem.e responses to econOlnic system.s, attitude

    Denm.ark and New Zealand*percent o f respondents)

    Denmark

    Strongly Stronglydisagree gree

    One Twa One Twa

    t t i tude st tement stage st geb

    stage stage Denmark (New Zealand) shouldmove towards socialism 6.1 2.0 .15 69.4 81.6 .05The private enterprise system is best for smallbusiness firms rather than big business firms 0.0 18.6 .01 14.0 23.3 .10 The private enterprise economic system issynonymous with small business 11.6 27.9 .0 1 9.3 30.2 .01In order to achieve economic growthenvironmental laws should be relaxed 0.0 8.3 .01 30.6 70.8 .01Capitalism does not encourage small business 2.1 2.2 .97 56.3 71.7 .03Big business operates best within a frameworkof capitalism rather than private enterprise 0.0 12.2 .01 2.4 26.8 .01 The economic system of capitalism is

    ynonymous with big business 0.0 19.6 .01 19.1 43.5 .01Capitali sm is socially equitable 0.0 13.3 .01 8.9 26.7 .01Socialism leads to high living standards 0.0 4.2 .04 59.6 64.6 .48Capitalism and private enterprise aredifferent economic systems 2.3 9.1 .04 16 .3 54.5 .01Communism is efficient in allocatingresources 2.1 4.3 .38 55.3 74.5 .01Government should guarantee a

    good standard of living 2.0 28.3 .01 10.2 28.3 .01Private enterprise is a most suitable economicsystem for different -sized business firmsthan is capitalism 0.0 5.7 .02 5.6 2.9 .35

    *Excludes 'don't know', and 'no opinion'. aN =49 bN =49 eN =60 dN =61

  • 8/12/2019 36 the Likert Scale Revisited

    9/18

    Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

    A I b 3 m r h e L i k t r r s e l r t \ i s i t c d : a n a 1 e r n a t t \ e r s i n n

    strongly disagree to the one-stage version. The analysis was of most intense (i.e.extreme scale position), values only, i.e. strongly agree and strongly disagree .

    For Study 1 the absolute values of the proportion of subjects in each sample

    who responded strongly agree or strongly disagree varied across scales andsamples, as shown in tables 1 and 2. For both United States sampl es and that inNew Zealand, on an absolute basis without ex eption there were a greater numberof extreme position (most intense) responses (agree and disagree) from the twostage format. Another possible typ e of analysis would be to compute meanvalue s for each version and then test for significant differences using the Student

    test. Mean values tend to hide differences such as those for extreme intensityvalues . In fact, for Study I the number of differences within three of the foursample groups were within chance expectations based on the binomial

    distribution. Similar results emerged for the Mann-Whitney nonparametic testfor distribution. Our concern is with a scale format s ability to detect people withextreme intensity responses. In Denmark, there were only two instances, oneeach for agree and disagree , where the one-stage generated a greater responsefrequency than did the two-stage format.

    Turning now to Study 2, as stated earlier the general topic of this study wasconsumer and social alienation. A total of 37 Likert scales (see table 3) werepresented to the sample of business students in New Zealand. The basic findingsare presented in table 4. As shown , the absolute values varied across scales andscale types. For all but one of the 74 possible comparisons there were a greaternumber of extreme position (intensity) responses from the two-stage format.

    Table 3Scales used foral ienat ion s tud y

    1 Most companies are responsive to the demands of theconsumer.

    2. I t seems wasteful for so many companies to produce thesame basic product.

    3. Unethical practices are widespread throughout business.4. Retail stores do not care why people buy their products just

    as long as they make a profit.5. Shopping is usually a pleasant experience.6. People are unable to help determine what products will be

    sold in the store.7 . Advertising and promotional costs unnecessarily raise the

    price the consumer has to pay for a product.8. What a product claims to do and what it actually does are

    two different things.9. Mass production has done away with unique products.

    10. Misrepresentation of product features is just something wehave to live with.

    11. Harmful characteristics of a product are often kept from theconsumer.

    12.I t is

    embarrassing to bring a purchase back to a retail store.13. I tend to spend more than I should just to impress my friendswith how much I have.

    14. Even with so much advertising it is difficult to know whatbrand is the best.

    339

  • 8/12/2019 36 the Likert Scale Revisited

    10/18

    Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

    J 0 u r a 1 0 f t h e \ a r k c t e s e a r c h S ) C i c t y Vo l U ll C 3 N u m b c r A p r i l 9 9 7

    15. A sale is not really a bargain but a way to draw people into aretail store .

    16. t is difficult to identify with current trends and fads infashion.

    17. I often feel guilty for buying so many unnecessary products.18 Most brands are the same with just different names and

    labels .19. A product will usually break down as soon as the warranty is

    up.20. Business s responsible for unnecessarily depleting our

    natural resources.21. It s difficult to identify with business practices today.22. One must be willing to tolerate poor service from most retail

    stores.23. t is difficult to know what store has the best buy.24. Business s prime objective is to make money rather than

    satisfy the consumer.25 . I often feel frustrated when I fail to find what I want in the

    store.26. After making a purchase I often find myself wondering why.27. It is hard to understand why some brands are rwice as

    expensive as others.28. t s not unusual to find out that business had lied to the

    public.29. Buying beyond one s means is justifiable through the use of

    credit.30. t s often difficult to understand the real meaning of most

    advertisements.31. Products are designed to wear out long before they should.32. Most claims of product quality are true.33. I am often dissatisfied with a recent purchase.34. The wide variety of competing products makes intelligent

    buying decisions more difficult.35. Advertisements usually present a true picture of the product.36 . A product S price will usually reflect its level of quality.37. What other people think will often influence what I buy.

    For Study 3 the absolute values of the proportion of subjects who respondedstrongly disagree varied widely across scales, as shown in table 5. Similar

    variation occurred for subjects responding strongly agree in the two-stagetreatment condition. However, there was little variation for strongly agree inthe one-stage condition. Of the 34 possible comparisons there was only onewhere the one-stage version elicited a greater extreme intensity response thandid the two-stage format.

    i s c u s s io n a n d conclusionsOn the basis of the three studies reported in this paper an overwhelmingconclusion is that the two-stage Likert scale elicited greater extreme positionvalues than did the one-stage format. Although not all proportion differenceswere statistically significant the vast majority were at

  • 8/12/2019 36 the Likert Scale Revisited

    11/18

    Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

    A l b a u m T h e L i k e r t s c a l e r e v i s i t e d a n l t e r n t e v e r S i o n

    Table 4 Scrmtgly disagree Scrongly agreeExtreIne responses to One- Two- One- Two-

    Scale stage stage< p stage Stage< palienation scales

    Percent o f respondents) I. 7.8 20.6 .01 4.9 12.3 .022. 22.3 50.7 .0 1 3.9 6 .9 .223. 3.9 5.5 .48 19.4 34 .2 .014. 2.9 8.2 .03 13.6 41.1 .015. 6.8 19.2 .01 8.7 27.4 .016. 13 .6 28.8 .01 3.9 15.1 .017. 12.6 30.1 .01 4.9 15.1 .018. 1.9 8.2 .01 10.7 21.9 .019. 12.6 26.0 .01 9.7 11.0 .69

    10. 21.4 56.1 .01 5.8 11.0 .08

    II. 2.9 5.5 .23 12.6 42.5 .0112. 25 .2 32.9 .11 7 .8 19.2 .0113. 40.8 60.3 .01 1.9 4.1 .2314. 3.9 1.4 .15 18 .4 50.7 .0 115. 2 .9 4.1 .55 27.2 49.3 .0116. 5.8 28.8 .01 2.9 8.2 .0317. 14.6 19.2 .25 3.9 12.3 .0118. 6 .8 12.3 .08 4.8 24.7 .0119. 12.6 26.0 .01 1.9 13.7 .0120. 13.6 32.9 .01 3.9 11.0 .02

    2l . 7.8 26.0 .01 2.9 4 .1 .5422. 40.8 71.2 .01 1.9 13.7 .0123. 3.9 9.6 .04 11.6 34.2 .0124. 6.8 11.0 .17 16.5 34.2 .0125. 1.0 26.2 .01 26.2 67.1 .0126. 15.5 30.1 .01 2.9 11.0 .0127. 9.7 15.1 .13 10.7 37 .0 .0 128 . 4.8 6 .8 .43 7.8 26.0 .0129. 33.0 52.0 .01 1.9 11.0 .0130. 9 .7 28.8 .01 5.8 11.0 .0831. 4.8 11.0 .03 10.7 24.7 .0132. 10.7 20.5 .02 1.0 9.6 .0133 . 5.8 23.2 .01 2.9 9.6 .0134. 10.7 13.7 .39 5.8 27.4 .0135. 13.6 27.4 .01 0.0 1.4 1 236. 13.6 20.5 .09 3.9 24.7 .0137. 9.7 13.7 .25 1.9 15.1 .01

    aSee table 3 =103 cN =73

    exceeded chance binomial probabilities. In light of this finding it appears thatLikert scales as generally used tend to underestimate the extreme positions heldby people, and that a central tenden(v forms-related error exists. Perhaps whenfaced with the standard one-stage format respondents are reluctant to express anextreme position even though they have it. The two-stage would give them theflexibility to express their true opinions. It is always possible that what has been

    341

  • 8/12/2019 36 the Likert Scale Revisited

    12/18

    Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

    J U r a I t f t h e :\ a r k e t R e s e d r : h S o c i e t y V t I u m e 3 q K u m b e r p r i l 1 9 9 7

    Table 5 Strongly dIsagree Strongly agreeMost intensity responses One Two One TwoScale item stage > stage< stage > stage