magnitude <strong>of</strong> the mode coefficients were highenough to be <strong>of</strong> practical importance towarrant attention. Finally, we applied poststratificationweights at the student-level forall survey items to minimize nonresponse biasrelated to sex <strong>and</strong> enrollment status.We analyzed the Web-only <strong>and</strong> Web-optionresults separately against paper as shown inTable 5 by Model 1 (Web-only) <strong>and</strong> Model 2(Web-option) against paper. We comparedWeb-only against Web-option in Model 3.For 39 <strong>of</strong> the 67 items, the unst<strong>and</strong>ardizedcoefficients for Model 1 favored Web-onlyover paper. For Model 2, 40 <strong>of</strong> the 67 itemsshowed statistically significant effectsfavoring the Web option over paper. Incontrast, there are only 9 statisticallysignificant coefficients that are morefavorable for paper over Web in Models 1 <strong>and</strong>2 combined. Model 3 reveals that there arerelatively few statistically significantdifferences between the two Web-basedmodes.The effect sizes for most comparisons in bothModel 1 <strong>and</strong> Model 2 are not large --generally .15 or less, with a few exceptions.Interestingly, the largest effect sizes favoringWeb over paper were for the three computerrelateditems: “used e-mail to communicatewith an instructor” (EMAIL), “used anelectronic medium to discuss <strong>of</strong> complete anassignment” (ITACADEM), <strong>and</strong> self-reportedgains in “using computers <strong>and</strong> informationtechnology” GNCMPTS).These models take into account many student<strong>and</strong> school characteristics. However, theresults for items related to computing <strong>and</strong>information technology might differ if a moredirect measure <strong>of</strong> computing technology atparticular campuses was available. That is,what appears to be a mode effect mightinstead be due to a preponderance <strong>of</strong> Webrespondents from highly Awired@ campusesthat are, in fact, exposed to a greater array <strong>of</strong>computing <strong>and</strong> information technology.On balance, responses <strong>of</strong> college students toNSSE 2000 Web <strong>and</strong> paper surveys showsmall but consistent differences that favor theWeb. These findings, especially for itemsunrelated to computing <strong>and</strong> informationtechnology, generally dovetail with studies insingle postsecondary settings (Layne,DeCrist<strong>of</strong>oro, & McGinty, 1999; Olsen,Wygant, & Brown, 1999; Tomsic, Hendel, &Matross, 2000). This said, it may bepremature to conclude that survey modeshapes college students= responses. First,while the responses slightly favor Web overpaper on a majority <strong>of</strong> items, the differencesare relatively small. Second, only itemsrelated to computing <strong>and</strong> informationtechnology exhibited some <strong>of</strong> the largesteffects favoring Web. Finally, for specificpopulations <strong>of</strong> students mode may havedifferent effects than those observed here.In auxiliary multivariate analyses, we foundlittle evidence for mode-age (net <strong>of</strong>differential experiences <strong>and</strong> expectationsattributable to year in school) or mode-sexinteractions, suggesting that mode effects arenot shaped uniquely by either <strong>of</strong> thesecharacteristics.Additional information about the analysis <strong>of</strong>mode effects is available in the NSSE 2000Norms report (Kuh, Hayek et al., 2001) <strong>and</strong>from Carini, Hayek, Kuh, Kennedy <strong>and</strong>Ouimet (in press). A copy <strong>of</strong> the Carini et al.paper can is on the NSSE website. We willcontinue to analyze NSSE data in future yearsto learn more about any possible mode effects.<strong>Framework</strong> & <strong>Psychometric</strong> <strong>Properties</strong>Page 18 <strong>of</strong> 26
Table 5: REGRESSIONS OF ENGAGEMENT ITEMS ON MODE OF ADMINISTRATION AND SELECTEDSTUDENT AND INSTITUTIONAL CONTROLS a,b,cModel 1:Web-only vs. PaperModel 2:Web-option vs. PaperModel 3:Webonlyvs. Web-optionItemUnst<strong>and</strong>ardizedCoefficientE.S. dUnst<strong>and</strong>ardizedCoefficientE.S.Unst<strong>and</strong>ardizedCoefficientE.S.CLQUEST .066*** .08 .053*** .06 .013 NSEMAIL .251*** .25 .151*** .15 .100*** .11CLPRESEN .063*** .07 .041*** .05 .022 NSREWROPAP -.026 NS . 025 NS -.051*** -.05CLUNPREP .096*** .15 .071*** .11 .025 NSCLASSGRP .196*** .24 .163*** .20 .033 NSOCCGRP .155*** .18 .083*** .09 .072*** .08TUTOR .097*** .12 .089*** .11 .008 NSCOMMPROJ .061*** .08 .040*** .05 .021 NSITACADEM .318*** .32 .194*** .20 .124*** .12FACGRADE -.015 NS .043*** .05 -.059*** -.07FACPLANS .038*** .04 .049*** .06 -.011 NSFACIDEAS .038*** .05 .076*** .10 -.038 NSFACFEED .029 NS .037*** .05 -.008 NSWORKHARD -.010 NS -.024 NS -.014 NSFACRESCH .054*** .07 .045*** .06 .009 NSFACOTHER .034*** .04 .021 NS .014 NSOOCIDEAS -.048*** -.06 -.063*** -.07 .014 NSDIFFSTUD .072*** .08 .051*** .05 .021 NSDIVRSTUD .040 NS .045*** .05 -.005 NSREADASGN f .062 NS -.047 NS .109 NSREADOWN f .405*** .09 .367*** .08 .038 NSWRITEMOR f .328*** .09 .101 NS .227*** .06WRITEFEW f -.067 NS .286*** .04 .353*** .05EXAMS .035 NS .100*** .06 -.065 NSMEMORIZE .036 .04 .032 NS .003 NSANALYZE .059*** .07 .045*** .05 .014 NSSYNTHESZ .083*** .09 .077*** .08 .006 NSEVALUATE .087*** .09 .114*** .12 -.027 NSAPPLYING .072*** .08 .079*** .08 -.007 NSACADPREP f -.737*** -.09 -1.228*** -.15 .491*** .06WORKON f .041 NS .305*** .05 -.264 NSWORKOFF f -1.368*** -.12 -.696*** -.06 -.673*** -.07COCURRIC f .667*** .11 .241 NS .426*** .06SOCIAL f .052 NS .383*** .05 -.331 NSCAREDEPD f -.258 NS .094 NS -.352*** -.05***p.001)f Metric derived from midpoints <strong>of</strong> response intervals, e.g., number <strong>of</strong> books read, papers written, or hours per weekg Factor change from logistic regression for dichotomous item (1=Yes, 0=No, “Undecided”=missing)<strong>Framework</strong> & <strong>Psychometric</strong> <strong>Properties</strong>Page 19 <strong>of</strong> 26