Lg.‘ .9®W¢tflk«5lcl .I\-I ..i)c¢vlli. e :3!!! ‘.!6 o . ‘6 A .3: .D.!A .9... I: .. I A p.31»; . v. 1. 79:... .1 . V . , .. . . .. . t.lll.y0.ln|lv..v ‘ .32::: £91. . . , . , .131, ivbivl~tdfrull {0.7.0:2) ‘ , .fi f1"!..n. vn. . gist; zei; .S.:F’v.>i.1.|. .. ‘ . . c .. I... .i‘hl‘flg: my...» in”. .IPoufzfl .1. , . r #x. 95)! :51... 1:5 3...... . ll; :2. 5.2.!!! w 1 . .15 ‘35.. 2 .1!» .I v .03.... . 710‘ If? |\.&!<.l '- $.3vo m. 3:53:11L.3 . 2:]... . x . . : ‘ .~ 5 7.1:... 54.331.153.15... 3.-.} .‘I. l’!»p’iI1I‘1l.3 17.. . s... 1.3!? .t\.l .1 11.1.1 ... 7v 3.... rtsrftl. . I'la‘snl.‘ ii“... :1: .. I 155..) .1: .. 3:3 uxwt..:}:7iv 25‘... t :s.v.\.i..t.7 I. III: £97.; 4 . :3 x 33.. .v. 35...! .‘wv’ liz.>i.$~\4 I I 3.1112,!)1 .‘ufi‘lifii Ila 1,.lt....2> l. t. ,. .- h ‘3‘)... in Y r ,{f:. ‘ . . . . . ‘ (‘IfquVAII..l.|0!I.b¢I¢rlplirI. pol». El I... a:vl..-Oll)....lvvlnl§.l.f§ . , . . 113:9)!!‘07. Irv-3.; lf.l.1ll[b’vy I! It“ r O . i 1 6: sun: 14"an (1.1.0:!Il3‘ctb .u»!-§~.:’!l..£3..z j . . . ‘ , .. t... .loltb-vl v. :1! ‘OI 1.7r‘lvlv- . (r. p"- . 4‘ u 1... I . x 4 . nil u... _ .1831, 5.1.. 3... 1:1,. ,. w 0* v .j. . {I‘VE-5.5.51. . ‘-..I;. in»... Cpl n.....f.. «er... I x .. . . .. I In ‘ .uoi‘n . :‘n v.1. v . I ‘ u . . - . II . ‘ O . ‘ . ‘ A I - I‘I. ‘ 4.. 5... . . ._ 1.: -I.I;l. .. . n . J . .w I." . : to: I. ‘ . . . - - ‘ , - It I; ‘ at x , .. , . . .. . A) . nfx..irF:a.fi:§84€.nh>ufl.?5: %.vn..r... ‘ L :i Wu 1 I. . £141.... afiamxfiakufigg V unnamnwxfiafiéxfi 4. , riéiztt‘9.2.5... J {3 .7... 2 . . it . ‘ v It. 35%. . I. ! 5...? 53.11.3125 5.32:2..1 .1. 21.. .. $7 1...? . , . .f. I . .-&u 5.1.7:}... .. i...p L1 , 1 . E... || II II I -f? I' :I -‘~vlu I HIG T Illmljlll £193 militia: 14 1122 This is to certify that the dissertation entitled PRE-SCREENING OF EMPLOYMENT CANDIDATES: A STUDY OF HOSPITALITY RECRUITER DECISION STRATEGIES presented by Michael P. Sciarini has been accepted towards fulfillment of the requirements for Ph . D . degree in Educational Administration [an 4mg Major professor Date March 8. 1993 MSU is an Affirmative Action/Equal Opportunity Institution O~ 12771 PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DliE DATE DUE DATE DUE J l MSUiIs An Affirmative Action/Equal Opportunity Institution cmmui PRE-SCREENING OF EMPLOYMENT CANDIDATES: A STUDY OF HOSPITALITY RECRUITER DECISION STRATEGIES BY Michael Peter Sciarini A DISSERTATION Submitted to: Michigan State University in partial fulfillment of the requirement for the degree of DOCTOR OF PHILOSOPHY Department of Educational Administration 1993 ABSTRACT PRE-SCREENING OF EMPLOYMENT CANDIDATES: A STUDY OF HOSPITALITY RECRUITER DECISION STRATEGIES BY Michael Peter Sciarini Purpose: Recognizing that the recruitment process may be complex and lengthy, and is usually made up of a series of smaller steps (e.g. pre-screening, on campus interviewing, post campus/second interviewing etc.), this study was undertaken to gain a better understanding of the decision making strategies and factors that hospitality recruiters utilize during the pre-screening step of the recruitment process. Methods: The Hospitality Recruiting Questionnaire (HRQ) was administered to 96 hospitality recruiters in attendance at the Michigan State University School of Hotel, Restaurant and Institutional Management's Career Expo XIV on November 17, 1992. A policy capturing analysis was conducted based upon data obtained from the recruiters' objective ratings of 40 employment candidate profiles (consisting of ten decision factors such as the candidate's grade point average, work experience, willingness to relocate etc.) and their subjective weighting of the same ten decision factors included in the HRQ. Results were analyzed using T-tests of mean differences as well as multivariate analysis of variance and Duncan's multiple range test. Findings for this sample of hospitality recruiters: 1. There were significant differences between the recruiter mean objective and subjective decision factor weights for five of the ten.decision factors considered in the study. 2. There were significant differences among recruiters from different hospitality segments with respect to their overall mean objective decision factor usage. 3. There were no significant differences among recruiters from different hospitality industry segments with respect.to their overall mean subjective decision factor usage when making pre-screening decisions. 4. The study revealed that individual recruiter differences (i.e. age” gender, experience, training, etc.) had no significant impact upon either objective or subjective decision factor usage for the recruiters in this sample. 5. The researcher also found relatively large standard deviations for most of the decision factors included in the candidate profiles. large standard deviations indicated a wide variation in factor usage by the recruiters in this sample. Given. these findings, the :researcher' concludes 'with recommendations for hospitality recruiters, students and faculty as well as suggestions for further research. Copyright by Michael Peter Sciarini 1993 ACKNOWLEDGEMENTS A large debt of gratitude goes first and foremost to the hospitality industry recruiters who responded to the questionnaire upon which this study was based. I would also like to extend my gratitude to the members of my committee: Howard Hickey, Bob Woods, Cas Heilman and Bill Cole. Eachtof these individuals was a resource (and never a roadblock) during this process - for which.I will always be grateful. Special thanks go to Phil Gardner of the Collegiate Employment Research Institute. Without his statistical expertise and coaching skills I could not have completed this study. Roberta Spaulding was patient, fast and competent in assisting in the manuscript preparation - Thanks Bert! And finally, to my wife Sue and our children Maria and Noelle, your patience and love have meant more to me than anything or anyone. The completion of a project such as this study provides an excellent opportunity to stop and express the gratitude that is so often unspoken on a day-to-day basis. Thank you and I love you. TABLE OF CONTENTS Chapter Page I. THE PROBLMOOOOOOOOOOOOOOOOOOOOO0.. 1 Introduction....................... 1 Purpose of the Study............... 2 Importance of the Study............ 2 Research Objectives................ 4 Research Hypotheses................ 5 Generalizability................... 6 Definition of Important Terms...... 6 summarYOOOOOOOOOOOOOOOOOOOOOOOOOOOO 8 II. REVIEW OF SELECTED LITERATURE...... 10 IntrOductj-onOOOOOOOOOOOOOOOOO0.0... 10 Characteristics Recruiters Consider Important (General).......................... 10 Characteristics Recruiters Consider Important (Hospitality Specific)............. 11 Decision Making in Recruiting...... 12 Policy Capturing................... 14 Summary............................ 23 III. RESEARCH METHODS................... 24 Introduction....................... 24 Research Design.................... 24 Research Procedures................ 27 Development of the Instrument...... 29 Data Collection.................... 34 Summary............................ 37 IV. FINDINGS.... ..... .................. 38 Introduction....................... 38 Description of Respondents......... 38 Recruiter Objective Decision Strategies......................... 42 Recruiter Subjective Decision Strategies......................... 43 vi TABLE OF CONTENTS (CONTINUED) Chapter Page Objective (Statistical) Weights vs. Subjective (Self-Report) Weights............................ 45 Recruiter Rating Consistency....... 48 Decision Factor Importance by Industry Segment................... 50 Decision Factor Importance by Individual Recruiter Differences... 55 Summary............................ 57 V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS................ 58 Introduction........... ............ 58 summarYOOOOOOOOOOOOOOOOOOOOOOOOIOOO 58 conCIUSions O O O O O O O O O O O O O O O O O O O O O O O I 62 Recommendat ions 0 O O O O O O O O O O O O O O O O O O O 7 1 BIBLIOGRAPHY. ...................... 74 APPENDIX A ........................ 82 Zedeck and Cascio (1982) Algorithm Hospitality Recruiting Questionnaire Survey Cover Letter APPENDIX B......................... 107 Individual Recruiter Policies Spearman Rank-Order Correlation Between the Statistical and Subjective Weights Individual Recruiter Multiple Ila From the Policy Capturing Analysis vii Table 10. 11. 12. LIST OF TABLES Page RESPONDENTS DESCRIPTIONS CLASSIFIED BY RECRUITING EXPERIENCE, HOSPITALITY OPERATIONS EXPERIENCE AND HOURS OF RECRUITING TRAINING RECEIVED................. 39 RESPONDENTS CLASSIFIED BY INDUSTRY SEGMENTOOCOOCOOOOCOOCOOCOOCOOCOCCOCO00......O. 4o RESPONDENTS CLASSIFIED BY ORGANIZATION SIZE.. 41 LEVEL OF INPUT IN THE HIRING PROCESS......... 41 MEANS OF OBJECTIVE WEIGHTS BY DECISION FACTOROOOOOO...OOOOOOOOOCOOOOOOOOOI00......0.. 43 MEANS OF SUBJECTIVE WEIGHTS BY DECISION FACTOR.00.0.0...COOOOOOOOOOOOOOOOO0.0.0.0...O 44 FREQUENCY DISTRIBUTION OF THE SPEARMAN RANK-ORDER CORRELATION BETWEEN RELATIVE AND SUBJECTIVEWEIGHTSOOOOOOOOOOOO0.0.0.0.... 46 T-TEST BETWEEN THE OBJECTIVE AND SUBJECTIVE WEIGHTS BY FACTOR.‘OI.OOOOOOOOOCCOCOOOOOOOOOOO 48 FREQUENCY DISTRIBUTION OF THE R2 VALUES OBTAINED FROM THE POLICY CAPTURING ANALYSIS... 49 RELATIONSHIP BETWEEN R2 VALUES AND SELF-REPORT MEASURE OF CONSISTENCY........... 50 FACTOR IMPORTANCE BY HOSPITALITY INDUSTRY SEGMENTOOOOOOOOOOOOOOOOOOOCOOOOOOOOOOOOOOO... 52 RECRUITER BETA WEIGHT SIGNS FOR THE FACTOR OF CANDIDATE GENDER.OOOOOOOOOOOOOOOCOOOOOOOO. 65 viii I TH ROBL INTRODUCTION The question "who are we going to hire?" represents one of the ‘most significant. personnel related. decisions any organization faces. Considerable resources (effort, time, dollars) may be expended.in attempting to answer this.question (American Management Association, 1986; Martin and Raju, 1992). The nature of the contemporary hospitality industry (i.e. its hypercompetitive/overbuilt markets, service focus and employee turnover/retention concerns) intensifies and multiplies the importance of the recruitment process. For an undergraduate student enrolled in a hospitality management school, questions such as "who will.I work for upon graduation?" and "what are recruiters looking for?" are no less significant considerations. Seeking answers to these and related.questions will likely be the impetus for allocation of a large portion of the students time and effort and may result in considerable personal anxiety — especially as graduation nears. The importance of the recruitment process to hospitality recruiters and students is clear. What is less clear than the importance of the recruiting process (and deserving of additional consideration) is how hospitality industry college 2 recruiters make decisions during the recruitment process. PURPOSE OF THE STUDY Recognizing that the recruitment process may be complex and lengthy, and is usually made up of a series of smaller steps (e.g. pre-screening, on-campus interviewing, post campus / second interviewing etc.) , this study was undertaken to gain a better understanding of the decision making strategies and factors that hospitality recruiters utilize during the pre-screening step of the recruitment process. IMPORTANCE OF THE STUDY The results of this study are important to at least three groups: hospitality recruiters, students and educators. For hospitality recruiters, the pre-screening decisions they make have considerable impact on the short and long term viability and success of their organizations. The results of this study may assist them in expanding their understanding of "common practice" within their profession (i.e. external environmental assessment). Participation in this study also provides a chance for self-assessment and reflection for each recruiter, opportunities which often seem few and far between 3 in the fast paced and competitive environment in which they function. Additionally, implications for the training of hospitality recruiters may follow from this study's findings. This study also provides valuable information to hospitality students as they embark upon their career journeys in the hospitality industry. By knowing what recruiters value (at the pre-screening stage) students may better prepare, perhaps even at an earlier stage in their college programs, to position themselves appropriately for the jobs they covet. Therefore, findings may help students avoid getting eliminated at the pre-screening stage of the recruiting process. Finally, for administrators and faculty of hospitality management schools, findings of this study may hold implications for curriculum development and career counseling. If administrators and faculty better understand recruiter decision-making criteria and strategies, they can better assist students in learning and developing in the appropriate areas. Again, the pre-screening step is critical because it comes so early in the recruitment process. While the opportunity for an interview does not guarantee an employment offer, being excluded from interviewing consideration certainly prevents an offer (at least temporarily). 4 RESEARCH OBJECTIVES This study explored the process by which a sample of hospitality recruiters used specific pieces of information (i.e. decision factors - a candidate's grade point average, willingness to re-locate, etc.) to make pre-screening decisions for a set employment candidates. The specific objectives of this study were: 1. To examine the relationship between actual and perceived decision factor usage by hospitality industry college recruiters when making candidate interview decisions. 2. To determine the relative differences (if any) in decision factor importance across hospitality industry segments (i.e lodging/hotels, fast food restaurants, full service restaurants, contract/non-commercial foodserviceiand.others). 3. To determine if individual differences affect the decision factor usage of hospitality recruiters when making candidate interview decisions. 5 RESEARCH HYPOTHESES From these research objectives the following research hypotheses were drawn and examined: 1. There is significant difference between hospitality recruiter objective and subjective decision factor usage when making pre-screening decisions. 2. There is significant difference among recruiters representing different hospitality industry segments with respect to their objective decision factor usage when making pre-screening decisions. 3 . There is significant difference among recruiters representing different hospitality segments with respect to their subjective: decision factor 'usage 'when. making' pre- screening decisions. 4. There is significant difference in hospitality industry recruiter objective decision factor usage with respect to individual recruiter differences. 5. There is significant difference in hospitality industry recruiter subjective decision factor usage with respect to individual recruiter differences. GENERALIZABILITY One never knows with certainty whether a sample, even a random sample, is representative of a larger population (Glass and Stanley, 1970). Though this sample was not randomly drawn from the population of all hospitality recruiters, a technique which could.have increased confidence about generalizability, the researcher has reason to believe the sample is minimally representative of leading hospitality organizations in terms of the number of Michigan State University Hotel Restaurant and Institutional Management graduates hired annually (Follow-up IReports, Michigan. State ‘University’ Career Development and Placement Services, 1985-1991). The reader will be invited to decide, given the research procedures described later and the actual demographics of the sample, whether results may be generalized to some population relevant to him or her. DEFINITION OF IMPORTANT TERMS Important terms as they relate to this study are operationally defined as follows: CONTRACT I NON-COMMERCIAL FOODBERVICE COMPANY - An organ 1 2 at ion 7 which offers on-site foodservice to employees, customers, patients, students, etc. in locations such as non—foodservice business (i.e. factories and offices), schools, hospitals, prisons and military installations. DECISION FACTOR - A piece of candidate information/data used by a hospitality recruiter when making pre-screening decisions. FAST FOOD RESTAURANT COMPANY - An organization which owns, operates or franchises restaurants that offer a limited menu and either drive-up or walk-up counter service. LODGING/HOTEL COMPANY - An organization which owns, operates or franchises hotel/lodging establishments. FULL SERVICE RESTAURANT COMPANY - An organization which owns, operates or franchises restaurants which may include fine dining, casual/theme, cafeteria and family dining establishments offering table service (as opposed to drive up and/or walk-up service). HOSPITALITY INDUSTRY - Lodging and foodservice businesses that provide short-term or transitional lodging and/or food and beverage. HOSPITALITY RECRUITER - An employee of a hospitality organization whose responsibilities include the recruitment of new employees (and specifically hospitality management school graduates). HOSPITALITY MANAGEMENT SCHOOL - A college or university offering a 4 year baccalaureate degree which emphasize the 8 acquisition of knowledge, skills and values necessary for success in the management of hospitality businesses. OBJECTIVE DECISION - A hospitality recruiter's statistical pre-screening factor usage as measured by the multiple regression analysis employed in this study. PRE-SCREENING DECISION - The decision made by a hospitality recruiter when he or she chooses which applicants will be interviewed and which will not (typically based upon a review of the applicant's resume/employment application). SUBJECTIVE DECISION - A hospitality recruiter's perceived pre- screening factor usage as measured by the recruiter's distribution of 100 points among the ten decision factors considered in this study. SUMMARY This chapter provided an overview of the purpose of this study. The research objectives and hypotheses were presented along with a thesaurus of important terms related to the study. The remainder of the body of the dissertation will be presented in the following fashion: Chapter II includes a summary and analysis of previously published literature related to this study. Chapter III will serve to explain the specific research methodology employed in the study followed by the study's findings which will be presented in Chapter IV. 9 Finally, Chapter V includes a summary of the key findings of the study as well as conclusions (especially for the practice of the three groups most concerned with hospitality recruiting) and recommendations for further research. II, BEZIEW OE SELECTED LITERATQBE INTRODUCTION Literature has been reviewed in four areas related to this study. In the order in which they will be discussed, they are: 1) characteristics recruiters consider important (general), 2) characteristics recruiters consider important (hospitality specific), 3) decision making in recruiting and 4) policy capturing methodology. CHARACTERISTICS RECRUITERS CONSIDER IMPORTANT (GENERAL) One need not look long to find a number of recent studies devoted entirely or in part to discovering the individual characteristics or attributes recruiters (in general) consider important when evaluating employment candidates. Among the factors shown to have significant influence at the pre- screening stage are work experience (Gardner, Kozlowski and Broadus, 1988; Holley, Higgins and Speights, 1988), intelligencelgrades (Muchinsky and Harris, 1977; Posner, 1981), involvemenp in outsidezextra curgicular activities (Campion, 1978)and willingness to re-locate (Hanigan, 1991; Harcourt and Krizam, 1989). 10 11 CHARACTERISTICS RECRUITERS CONSIDER IMPORTANT (HOSPITALITY SPECIFIC) Considerably less seems to have been written about hospitality industry' specific studies of this sort. The hospitality - specific literature tends to focus on student expectations of such things as salary, promotions and.working conditions (Farmer & Tucker, 1989; McLeary & Weaver, 1988), and training programs (Durocher & Goodman, 1991). Research has also been conducted in such areas as student perceptions of their level of preparation to enter hospitality management positions (Knutson & ZPatton, 1992) as ‘well as criteria students utilize when choosing their first employment out of college (Knutson, 1987; Knutson, 1989; Laker and Gregory, 1989). Recent hospitality - specific studies have also examined the relevance of hOspitality curriculi (Hsu and Walsch, 1992; Walker, 1992). Recent studies have also explored the management competencies currently necessary for success in the hospitality industry (Cichy, 1991; Tas, 1988) as well as speculated upon what will be required of the hospitality manager in the year 2000 and beyond (National Restaurant Association, 1992). However, only one study’ examined hospitality recruiters and the criteria they use when evaluating employment candidates (Ley & Sandler, 1982). Their findings focused on 12 the importance of the actual interview and did not seek to effectively differentiate pre-screening (prior to the actual interview) decision factors from decision factors considered during the face-to-face contact between the recruiter and student applicant. Among the criteria the recruiter's in Ley and Sandler's study ranked as important, four potential pre- screening factors (work experience, willingness to re-locate, participation in extra curricular activities and grades) matched the criteria considered important by general business recruiters, as noted earlier. DECISION MAKING IN RECRUITING In addition to its small sample size (n=15), the Ley and Sandler (1982) study was impacted by a methodological shortcoming, one it shares with most of the general studies cited earlier. In these studies, recruiters were typically presented with or requested to create a list of criteria they look for in. applicants and. then are asked to rate the significance of each factor to candidate appeal (see Scheetz, 1991 for another example). The attributes were then either ranked according to importance or statistically tested for significance based on mean differences. Methodologies of this sort are limited in that they can only reflect the factors recruiters "think" (subjectively) are significant in their 13 candidate evaluations. These evaluations may, but also may pp; accurately demonstrate the criteria that recruiters actually use when rating applicants. Research has demonstrated that individuals are often not aware of the factors they use when reaching decisions (Nisbet, Krantz, Jepson & Kunda, 1983; Nisbet & Wilson, 1977); and this may be the case with hospitality recruiters. It has also been shown that decision makers are often unaware of how frequently they utilize certain factors when reaching conclusions (Huber; Northcraft.and Neale, 1990; Stahl & Zimmer, 1984). Stahl's and Zimmer's study (1984) demonstrated that business executives, though they believed they utilized a broad range of criteria in making decisions, actually considered a significantly smaller range of the available factors in their decision making practices. A recent (non-hospitality specific) study compared the factors recruiters .§§ig; they’ used and. the factors they actually utilized when making pre-screening decisions. Findings demonstrated that despite their belief that they were using a broad range of characteristics when evaluating candidates, recruiters actually relied primarily on only two factors (e.g. major grade point average and an artificial measure of communication skills) when making interview pre- screen decisions (Gardner, Kozlowski and Hults, 1988). The sample in the above study (recruiters seeking candidates in electrical engineering, computer science, 14 general business, accounting and the social sciences/liberal arts) resulted in large differences in decision strategies among the recruiters. Large standard deviations were also reported for all the candidate characteristics considered, indicating a wide variation in factor usage by this sample of recruiters. Though findings were likely impacted by its heterogeneous sample, the Gardner, Kozlowski, and Hults (1988) study provides a useful model, through its use of a technique known as policy capturing, to analyze recruiter decision making and overcome the methodological limitation of the earlier studies (a more complete discussion of the policy capturing technique follows). POLICY CAPTURING Policy capturing is a statistical technique used to address unanswered questions regarding decision making processes. The idea of using policy capturing to model how information is integrated into decision-making strategies was developed by Hoffman (1960). Hoffman suggested that mathematical models could be created which link specified stimulus information to judgmental outcomes through the development of a multiple regression equation based on a combination of judgements by an individual. This regression equation serves as a "paramorphic" description of the cognitive processes utilized by the rater. Hoffman used the 15 term paramorph (from the field of mineralogy) because while the mathematical representation of the judgement process does not directly model the thinking of the rater, it is analogous to the situation in mineralogy in which two minerals can have the same chemical composition, but different underlying molecular structures. The type of analysis used to analyze the different elements determines the different underlying structures. Policy capturing does not result in an exact representation of the judgement processes of the raters, but does model how they combine and integrate information on a given judgement. The process can then be inferred through analysis of both the stimulus information (input variables given to the rater) and the outcomes (the decisions actually made) (Dawes & Corrigan, 1974; Oskamp, 1967). Policy capturing represents decision-making at a general level while verbal protocols and policy tracing may represent the process at a more specific level (Einhorn, Kleinmuntz,& Kleinmuntz 1979; Ford et al, 1989). Policy capturing typically involves the following steps: 1) A number of profiles containing various characteristics (decision factors) of real or hypothetical situations or individuals are collected or created. 2) Raters are asked to study and analyze the information, resulting in a rating of each individual profile. 3) A multiple regression equation is computed for each rater by regressing the individual profile 16 ratings onto values associated with the decision factors contained in each profile. The regression results in the information necessary to evaluate the raters decision strategy. 4) After rating all profiles, raters are asked to state their rating policy which can then be compared to the objective rating policy obtained from the multiple regression equations. The profiles usually include a limited number of pieces of information or decision factors (usually 3-10) which may be represented as numerical or categorical responses. Policy capturing has been used to study decision-making strategies in a variety of situations utilizing decision factors such as workplace disciplinary action and job terminations (Klaas and Wheeler, 1990, Rousseau and Anton, 1991), corporate merger decisions (Feldman and Murata, 1991) , stock-market data (Ebert and Kruse, 1978) and even information concerning the safeguarding of nuclear power plants (Brady and Rappaport, 1973). The number of profiles in a given study is usually a function of the number of decision factors contained in each profile. As noted earlier, once the profiles are obtained or created, the rater is asked to analyze and interpret the characteristics contained in each profile into an overall assessment or judgement. Raters are typically asked to state their subjective rating policy after completing the objective rating task for all profiles. The subjective rating policy is the rater's 17 perception of the relative importance of the decision factors which were used when making the assessments or judgements. The technique for gathering the subjective weights usually requires raters to distribute 100 points among the decision factors such that the distribution represents the relative importance of each of those variables to the final decision (Hoffman,1960). The 100 point distribution has been the most commonly used, although other methods have been applied (Cook and Stewart, 1975; Doherty and Keely, 1972). The most significant step in the analysis of the results of a policy capturing study is the development of a multiple regression equation for each rater by regressing the overall rating onto the values of the decision factors contained in each profile. The resulting multiple R? is a measure of how well the decision factors account for the linear portion of the variance in the overall ratings, while the beta weights obtained for each factor paramorphically represent the weighting strategy used by the rater. A comparison may then be made between the "objective" (beta) weights from the multiple regression procedure and the "subjective" (stated) weights to analyze how well a rater's statistical policy matches the stated policy. There are two consistent findings with regard to a general review of policy capturing studies. First, the general linear model has performed well in describing rater policies, as evidenced by the consistently high rater R2 found 18 when regressing overall ratings on scores for the separate profile decision factors (Hobson and Gibson, 1983; Slovic and Lichenstein, 1971; Slovic et a1. , 1977) . Second, rater subjective policies tend to be dissimilar to their statistical or objective rating policies. It has typically been found that raters' subjective policies overestimate the number of statistically significant decision factors obtained from the regression analysis (Hobson and Gibson,1983; Gardner, Kozlowski & Hults, 1988). These findings are meaningful in demonstrating’ the efficacy' of applying' policy' capturing’ procedures to the analysis of recruiter pre-screening decisions. However, the policy capturing methodology is not without its problematic issues. One such problematic issue concerns the impact of the number of decision factors presented to the rater on the R2 of the rater's regression equation (the consistency with which raters weight and combine the decision factors in the overall judgement). The higher the R? obtained, the greater consistency of the rater. Studies on this topic have generated mixed results. Einhorn (1971) compared subjects' R2 obtained using two, four, or six decision factors. He found that the values of R2 increased with fewer numbers of decision factors. In addition, subjects reported that they felt the task was more difficult with an increasing number of decision factors. Cook and Stewart (1975) and Billings and Marcus 19 (1983) also found higher R2 with fewer decision factors, as compared to decisions with a greater number of factors. However, Anderson (1977) compared R2 for tasks involving four, six or eight decision factors and found no differences across conditions. Although no clear conclusions can be drawn from this research, one would expect that the R2 would be higher given a small number of decision factors, since such decisions should be less complex than those with a greater number of factors. Large numbers of decision factors may overload our information processing capacity. Thus, it would be easier to weight or combine a small number of decision factors in a consistent fashion. When a larger number of decision factors are presented, subjects may cognitively reduce the set of factors to a more meaningful number so that they are better able to process this information (Miller, 1956). This suggestion has been borne out in the research presented earlier in which the researchers found that a small subset of the presented decision factors usually account for large proportions of the variance in raters' judgements (e.g. Gardner, Kozlowski and Hults, 1988). A related issue concerns the level of intercorrelation among the decision factors presented to the subject raters. Schenk and Naylor (1968) showed that as the amount of decision factor intercorrelation increases, subjects' responses become more systematically a linear function. Stated differently, as 20 the factor intercorrelations increase, the R? for each rater should increase accordingly, solely on the basis of this statistical artifact. Due to Schenk and Naylor's suggestions, research using policy capturing has typically constrained the intercorrelations between decision factors to be zero, which may or may not accurately reflect reality (Hobson and Gibson, 1983; Schmitt and Levine, 1977). A third issue relating to the nature of decision factors used in policy capturing tasks is that of format. Anderson (1977) compared factors presented in verbal/paragraph form with factors presented. in. numerical form. IHer results indicated that subjects were more consistent in their ratings when rating numerical factors thanwwhen.rating verbal factors. One possible explanation for this finding may have been that raters were not able to unambiguously interpret the verbal paragraph cues. Research concerned with the issue of discovering whether subject raters combine decision factors in a linear or non- linear fashion has yielded conflicting results. Einhorn (1971) found that non-linear models outperformed the linear models, while Goldberg (1971) and Ogilvie and Schmitt (1979) found that linear models outperformed non-linear models. As Goldberg (1971) suggested, there were important differences in the nature of these studies which could have resulted in the conflicting conclusions. These differences include: the kind of judges, the type of task, the number of decision factors, 21 the intercorrelations among the decision factors, type of responses required (rating vs. ranking), values for decision factors (discrete vs. continuous) and the number of profiles being evaluated. A number of these factors relate to the previous discussion regarding the nature of the decision factors utilized, while others draw attention to additional factors which might be relevant to the rating process. The notion that the type of task or the nature of the decision required can influence whether raters use decision factors in a linear or non-linear fashion is an important one. This notion is consistent with research done using policy-tracing (verbal protocol) procedures which show that rating tasks require different kinds of cognitive processing than choice/preference tasks, in that the latter require more configural use of decision factors than do the former (Billings and Marcus, 1983; Payne, Braunstein, and Carroll, 1978; Svenson, 1979). Another problematic issue concerns the relationship between subjective or stated rating policies and the objective or statistical rating policies. Following Hoffman's (1960) suggestions, most of the previous studies using policy capturing procedures have used the method of asking subjects to distribute 100 points among the decision factors, in order to obtain subjects' subjective weighting of the dimensions. Using the statistical weights that contribute significantly to the regression equation as the measure of objective factor 22 usage, results from these studies have consistently indicated that subjects overestimate the number of factors that they actually use in making their judgements. Questions have been raised as to*whether this method allows raters the opportunity to state that they are using factors in a non-linear fashion. Cook and Stewart (1975) addressed this issue through comparison of seven different techniques for obtaining subjective weights. They compared the traditional method with both additional linear and non-linear methods and found that there were no major differences between the methods. The authors concluded that the 100 point allocation method was as good as any other, and recommended its continued use, primarily because it is the simplest method to use. A final issue concerns whether subjects should state their policies before or after completion of the rating task. Balzar, Rohrbaugh, and Murphy (1983) found that subjects who completed the rating task first had significantly higher reliabilities for their predictions based on their subjective policies than did subjects who completed their subjective policies before completing the rating task. The authors hypothesized that this result was due to raters having an opportunity to monitor their decision behavior and assess their strategies "in practice" when giving their subjective weighting policies after making the ratings. 23 SUMMARY Literature relevant to this study of hospitality recruiter employment pre-screening decisions has been reviewed in four areas: (1) characteristics recruiters consider important (general), (2) characteristics recruiters consider important. (hospitality’ specific), (3) decision.:making' in recruiting and (4) policy capturing methodology. The importance of the employment pre-screening decision to recruiters and employment candidates is evident. Significant study of the characteristics recruiters (in general) consider important has occurred. Only one study was found which contributed any empirically derived information to the understanding of which characteristics hospitality recruiters consider important" Limitations to the methodologies employed in most of these studies limits their usefulness. Despite minor problematic issues, the efficacy of the use of the policy capturing methodology to overcome the limitations of the previous studies has been demonstrated. Therefor, given the paucity of hospitality specific studies of this sort and the overall effectiveness of the policy capturing methodology, this research employed policy capturing to study hospitality recruiter pre-screening decision strategies. Chapter III provides a description of the specific research design and methodology employed in this study. III. RESEARQE METHODS INTRODUCTION The purpose of this chapter is to describe the research design and methodology used in conducting this study. The policy capturing framework utilized by Gardner, Kozlowski and Hults in their 1988 study of general recruiter decision strategies was modified to develop a research instrument, the Hospitality Recruiting Questionnaire (see Appendix A) which was administered to 96 hospitality recruiters. These recruiters were attendees of the Michigan State University School of Hotel, Restaurant and Institutional Management's CAREER EXPO XIV, held on November 17, 1992. RESEARCH DESIGN The following research hypotheses were tested at the .05 level of significance: flypothesis 1 There is significant difference between hospitality recruiter objective and subjective decision factor usage when making pre-screening decisions. 24 25 Hypctgesis 2 There is significant difference among recruiters representing different hospitality industry segments with respect to their opjective decision factor usage when making pre-screening decisions. Hypothesis 3 There is significant difference among recruiters representing different hospitality industry segments with respect to their subjegtive decision factor usage when making pre-screening decisions. flypothesis 4 There is significant difference in hospitality industry recruiter objective decision factor usage with respect to individual recruiter differences. Hypothesis 5 There is significant difference in hospitality industry recruiter subjective decision factor usage with respect to individual recruiter differences. Research hypothesis 1 was tested using the t-test for differences between means, SPSS-X subprogram T-PAIRS (Hedderson, 1987). This technique was chosen because it allowed for a test of statistical significance on the 26 difference between each pair of mean decision weights (objective and subjective) for all ten. decision factors considered in this study. Research hypotheses 2 thru 5 were tested using multivariate analysis of variance, SPSS-X Subprogram MANOVA (Hedderson,1987). This technique was chosen because of the ability of the tests used to attend to the data as a whole rather than to each set of comparisons of means separately. Analysis of each of the measures separately results in redundancy, such that statistical error rates may be multiplied manifold. The multivariate model retains the multiple scores as a set of interrelated traits (Bray and Maxwell, 1985). SPSS-X Subprogram MANOVA provided multivariate analysis of variance for both sets of dependant variables -- objective decision weights and subjective decision weights -- for (1) the interaction effect between two of the independent variables, and then (2) for each of the independent variables while controlling for any effect of the other. Subprogram MANOVA then provided univariate analysis of variance for each dependent variable in the same fashion. The researcher was therefor able to test for the effect of each independent variable on objective and subjective decision weights, for the interaction effect between the independent variables (e.g. hospitality industry segments and recruiter individual differences), and for each effect and interaction on each set 27 of dependent variables. MANOVA.provided four criteria for tests of significance: Wilks Lambda, Hotelling's trace criterion, Roy's largest root criterion, and Pillai's criterion. Each of the four gives a single probability statement to retain or not retain the null hypothesis of no treatment effects. Where significant differences were found, the direction of those differences was determined by inspection. Where there were more than two levels, Duncan's multiple range test was utilized to determine between which levels the differences were significant. RESEARCH PROCEDURES The study had no treatment and one observation. Participants were administered the H_O§pitalitv Recruiting Quespionnaire (HRQ). The HRQ included three parts to which participants responded" Part.1 - Background, consisted of two sections, A, and B. Section A included 15 questions related to the respondents (1) previous recruiting experiences and current level of involvement 'with. reCruiting on college campuses; (2) previous hospitality industry operations experience; (3) amount of special training in recruiting; (4) individual demographics (e.g. age and gender); (5) organization (e.g. industry segment, number of units/properties and whether pre-screening decision criteria 28 are provided by the organization); and (6) level of input in the hiring decision using a five-point scale where one = "pre- screen only" ranging to five = "full hiring authority." Section B asked participants to list the factors they would ideally like to consider when reviewing hospitality student candidates for possible interviews. After listing the factors, respondents were asked to go back and rate the level of importance of each factor, using a five-point scale where one = "slightly important to me" ranging to five = "extremely important to me". Part 2 - Applicant Evaluation, asked each participant to evaluate the same set of 40 fictitious candidates for job interviews. Each of the potential candidates was represented by an individual credential profile which included ten pieces of information (decision factors) about that candidate. Before asking for completion of the evaluation task, the HRQ included a brief (less than two page) explanation of the profiles and decision factors to inform the participant of what was included in the profiles and how to interpret the decision factors (further discussion of the profiles follows in the Development of the Instrument section.) Part 3 - Information Use Questionnaire, consisted of three multi-part questions. Questions 1A, 1B and 1C asked the participants to use a five point scale (where 1 = "to a very minimal extent and ranging to 5 = "to a very great extent") to rate the manner in which they used the information provided in 29 the profiles. Question 1A asked the extent to which the participant believed they made gppsistent ratings, question 13 asked the extent to which they believed they had distinguished the applicants with their ratings (a measure of discrimination, as opposed to rating all applicants the same) and question 1C asked the extent to which they believed they were gygpg of and could accurately describe the factors that were important to their ratings. Question 2 asked the participant for a subjective impression of how they used the candidate information. Each recruiter distributed 100 points among the 10 cues included in each profile to obtain the subjective measure. The distribution of points indicated their stated importance of each factor to the decisions the participant had made. Question 3 asked for a yes or no response as to whether they believed there was anything unusual or odd about any of the applicants. Respondents selecting "yes" were invited to explain (in an open ended way) what they thought was unusual. DEVELOPMENT OF THE INSTRUMENT The pilot study instrument was derived from the questionnaire utilized by Gardner, Kozlowski and Hults in their 1988 study of general recruiter pre-screening decisions. The previously used instrument was modified by the researcher to make it suitable for this investigation of hospitality 30 recruiter pre-screening decisions. The most significant changes came in two areas. First, in Part I - section A, questions specifically related to hospitality organizations (lodging/hotel companies, full service restaurants, non- commercial foodservice companies, fast food restaurants and others) were added. A question asking the participant to rate their level of input in the hiring process was also added. Secondly in Part 2 - Applicant Evaluations, revisions were made to the cues included in the applicant profiles to be rated. Based.on the review'of related literature, an analysis of the student credential forms recruiters receive when interviewing on-campus (provided by the Michigan State University Career Development and Placement Services office) and a review of common undergraduate resumes as contained in the Michigan State University, School of Hotel, Restaurant and Institutional Management's annual resume book for the years 1987 thru 1992, ten decision factors were selected to create the fictitious candidate profiles used in this studyu The ten decision factors were (1) career objective, (2) overall grade point average, (3) major' grade point average, (4) work experience, (5) extracurricular activities, (6) percentage of education. paid for personally, (7) computer skills, (8) language skills, (9) willingness to re-lccate and (10) gender. The construction of the profiles proceeded as follows. 31 For each decision factor included in the profiles, three scale values representing dimensions of performance/achievement (i.e. low, average and high) were assigned. The profiles were then created through the use of an algorithm developed by Cascio and Zedeck (1982) such that the intercorrelations among the cues approximated zero and the mean scale values for each cue were 2.0 with a standard deviation of .78. The forty uncorrelated profiles were designed to ultimately ensure multiple regression equations which would yield interpretable beta weights (Gardner, Kozlowski and Hults, 1988) . Additionally, a concerted effort was made to create profiles which would be representative of and closely simulate actual candidates. Based on the literature review and other criteria used in the selection of cues, it is the researchers' opinion that the requirements of the statistical tools utilized did not detract from the "reality" of the profiles (Appendix A includes a copy of the Cascio and Zedeck (1982) algorithm). The forty candidate profiles were developed based on the algorithm, with a typical candidate format statement as follows: _CUES W 1)CAREER OBJECTIVE 2)OVERALL GPA 3)MAJOR GPA 4)EXTRACURRICULAR ACTIVITY 5)WORK EXPERIENCE 6)% OF EDUCATION PAID 7)COMPUTER SKILLS 8)LANGUAGE SKILLS 9)WILLINGNESS TO RELOCATE 10)GENDER OHNUNNHUNH 32 The resulting profile would then look like: BRIAN LEONARD: CAREER OBJECTIVE......OOOOOOOOOOOO0.....Broad OVERALLGPAOOOOOOOOO...OOOOOOOOOOOOOOOOZOO-2.6 MAJORGPAOOOOOOOOO00.00.00.000000000000304-400 WORK EXPERIENCED...OOOOOOOOOOOOOOO0.0'Cone LEVELIand Two LEVEL II EXTRA CURRICOACTIVITY......OOOOOOOOOOOONone % of EDUCATION PAID....................67% - 100% COMPUTER SKILLS............... ...... ...Average LANGUAGE SKILLS........................English plus conversant WILLINGNESS TO RELOCATE...............Very limited Gender was represented.by/imbedded in the first.names assigned to each profile. Twenty customarily male and twenty customarily female first names were assigned over the forty profiles (Gardner, Kozlowski and Hults, 1988). There was a deliberate.attempt.toiavoid.androgynous first.names (i.e. Pat, Kelly, Kim, Terry, Blair) so as to provide the study participant with.a clear notion of the gender of the candidate profile. Additionally, there was a deliberate attempt to avoid indicating the race or ethnicity of the candidates by the surnames selected for the profiles. Though race/ethnicity may make for an interesting variable in future research, it was deemed outside the scope of the present study. A pilot study/field test of the revised instrument was then undertaken to determine whether it was understandable and suitable for its intended purposes. IFour hospitality industry practitioners and/or educators were invited to participate in 33 the pilot study. The four individuals who participated were: Dr. Ronald F. Cichy, Director of the School of Hotel Restaurant and Institutional Management at Michigan State University. Dr. Cichy has a combination of over 20 years of hospitality business/education experience. Robert Poole, Jr. currently Assistant Director of Michigan State University's office of Career Development and Placement Services and former college recruiter for the Walt Disney World Company for three years. Angelos J. Vlahakis, Director of the School of Hotel, Restaurant and Institutional Management's Student and Industry Resource Center at Michigan State University, whose combined hospitality’ business and.teducation. experience exceeds 35 years. E. Allen Wetherell, Director of the W. A. Lettinga Business and Training Center of Davenport College, Grand Rapids, MI. Mr. Wetherell.has previously served as Director of Seminars and Conferences for the Educational Institute of the American Hotel and Motel Association. Additionally, Mr. Wetherell has over 15 years of combined hospitality business/ education experience. The feedback provided by the pilot participants suggested that the instrument appeared to be valid and appropriate. It was noted that the survey questions were clear and easy to follow and that the decision factors and levels/dimensions within each decision factor ‘were accurate and realistic 34 representations of typical undergraduate hospitality students. Two suggestions emerged from the pilot participants feedback. The first.was the notion that not all recruiters who came on campus have much input in the hiring decision, therefor a question was added to the instrument to attempt to get.a.measure as to the amount of input each individual had in the hiring process. Secondly, one pilot participant noted the lack of perceivably racially ethnic names and suggested this may be a variable to include. As was noted earlier, this variable was considered but not selected as it was deemed outside the scope of this study. DATA COLLECTION The 96 recruiters (representing 48 hospitality organizations) who attended the School of Hotel, Restaurant and Institutional Management's CAREER EXPO XIV on November 17, 1992 were invited to participate in the study. CAREER EXPO XIV consisted of a series of events including a luncheon and one half day of various presentations and seminars of interest to the recruiters and students in attendanceu The half day of presentations was followed by a dinner and a two and one half hour "job fair" intended for interested students to have an opportunity to meet with various hospitality recruiters in attendance. Each recruiter who expressed a commitment to attend 35 CAREER EXPO XIV received a letter on Michigan State University letterhead (dated September, 1992) signed by Ronald F. Cichy, Director of the School of Hotel, Restaurant and Institutional Management, briefly describing the study and inviting the recruiter to participate when on campus (see Appendix A). Additionally, during two different seminars/presentations made during CAREER EXPO XIV, specific mention of the HRQ was made and the recruiters in. attendance ‘were again invited. to participate in the study. At whatever point in the day each recruiter registered to attend the "job fair" portion of the event (a required activity), he/she received an 8 1/2" x 11" pocket folder which contained registration materials and the HRQ, prominently displayed on the top of one side of the folder. To draw the recruiter's eyes to the HRQ, fluorescent green paper was selected for use as the cover sheet. A self-addressed, postage paid return envelope was attached to the HRQ and the instructions advised participants of three options for return of the completed instrument. First, it could be dropped at the front desk of the Kellogg Center for Continuing Education (site of CAREER EXPO XIV). The second option was to drop the HRQ at the main desk of the Career Development and Placement Services offices (where most of the recruiters would be interviewing employment candidates on the day or days following CAREER EXPO XIV). The last option was to mail the HRQ in the postage paid envelope provided. The intention with providing these options was to 36 make it as convenient.as possible to participate in the study. No matter which return option‘was selected, a requested return date of November 30, 1992 was established in the instructions included in the HRQ. As further inducement to participate, recruiters were offered.a copy of the results of the study and the opportunity for their name to be included in a drawing to award an array of prizes. including a wristwatch, a sweater and a complimentary exhibit booth registration for the recruiter's organization at next year's CAREER EXPO. The drawing was to be held on December 1, 1992 and winners were to be notified by mail. By December 1, 1992, 23 recruiters had responded to the HRQ. Commencing on December 2, 1992 and continuing through December 18, 1992, the researcher conducted a follow-up telephone "campaign". Each non-responding recruiter was contacted by telephone and informed of the extension of the drawing date to December 30, 1992 and invited, again, to participate in the studyu By December 29, 1992, 49 recruiters had responded to the HRQ. On December 30, 1992 a mailing was sent to the remaining recruiters who had not yet responded, again inviting their participation and requesting a return date of January 13, 1993. By January 15, 1993 a total of 51 hospitality recruiters had returned the HRQ, a response rate of 53.1%. 37 SUMMARY Chapter III contained the research design and procedures utilized in the study including a description of the survey instrument and data collection techniques. The findings of the study are reported in Chapter IV. IV IN INGS INTRODUCTION Chapter IV presents the results of the study. The results are presented in tabular form where appropriate, accompanied by descriptions of significant components of the tables. DESCRIPTION OF THE RESPONDENTS The 51 respondents to the HRQ consisted of 35 males (68.6%) and 16 females (31.4%). The average age of the respondents was 36.6 years with a range from 22 to 67 years. The data in Table 1 describes the respondents in terms of their recruiting experience, years of hospitality industry operations experience and hours of recruiting and/or interviewing training received. 38 39 TABLE 1. RESPONDENTS DESCRIPTIONS CLASSIFIED BY RECRUITING EXPERIENCE, HOSPITALITY OPERATIONS EXPERIENCE AND HOURS OF RECRUITING TRAINING RECEIVED. I Mean Median Mode Range Times served as a recruiter 46.2 20.0 1 1 - 720 on college campuses Years involved in 5.6 4.0 <1 <1 - 20 recruiting on college campuses Years of hospitality 10.0 6.0 6.0 <1 - 35 industry operations experience Hours of special training 40.4. 40.0 40.0 0 - 175 in recruiting and/or interviewing Additionally, 28 recruiters (54.9%) reported that college recruiting is a full time or regular part of their job responsibilities. The remaining 23 respondents (45.1%) are occasionally asked to assist in college recruiting, but it is not one of their regular job responsibilities. Participants were asked to list their job titles. These titles were classified as either human-resources related (i.e. Recruiter, College Relations Manager, Human Resources Director) or operations related (i.e. Restaurant Manager, Vice President of Rooms Management, Owner-operator). Thirty-eight recruiters (74.5%) had human resources-related job titles while 13 recruiters (25.5%) had operations-related job titles. 40 Table 2 provides a description of the respondents in terms of the hospitality industry segment for which they recruit. TABLE 2. RESPONDENTS CLASSIFIED BY INDUSTRY SEGMENT. ( Number of Percent Industry Segment Responses of Total Lodging/Hotels 19 37.3% Fast Food Restaurants 3 5.9% Full Service Restaurants 12 23.5% Contract/Non-Commercial 12 23.5% Foodservice Operators *Other 5 9.8% *Included.specia1ty retail (1), executive search firms (3) and total resort destination(1). Six respondents (11.8%) reported they were employed by independent/single unit operations while 45 recruiters (88.2%) reported they worked for multi-unit organizations ranging from 2 to 9500 units. Table 3 includes additional data related to the size of the organizations represented by the respondents. 41 TABLE 3. RESPONDENTS CLASSIFIED BY ORGANIZATION SIZE. Number of Percent Number of Properties/Units Respondents of Total less than 50 16 31.4% between 50 and 200 12 23.5% between 201 and 600 16 31.4% greater than 600 units 7 13.7% Respondents were also asked whether their organization provided them with specific criteria to be used when deciding which employment candidates to interview. Thirty-five recruiters (68.6%) reported. that they‘ are ‘provided. with specific criteria while 16 participants (31.4%) reported their organizations do not provide specific pre-screening criteria. Recruiters also self-reported.their level of input.in the hiring process as presented in Table 4. TABLE 4. LEVEL OF INPUT IN THE HIRING PROCESS. Number of Percent Level of Input Responses of Total pre-screen only 2 3.9% minimal input 3 5.9% moderate input 7 13.7% significant input 26 51.0% full hiring authority 13 25.5% 42 RECRUITER OBJECTIVE DECISION STRATEGIES Assessment of the decision making strategies actually used by hospitality recruiters was one of the primary objectives of the study. Table 5 presents the means of the statistical / objective weights of the decision factors obtained by the policy capturing analysis (see Appendix B for individual rater policies). Recruiters relied largely upon the candidate's willingness to relocate (with a relative weight of 31.21) when.determining the candidate's suitability for an interview. Level of extra-curricular participation and work experience (with relative weights of 16.00 and 15.89, respectively) were the next most heavily weighted decision factors. The next most common factors used by the recruiters were grade point average in major courses (relative weight, 9.91), career objective (relative weight, 9.83) and overall grade point average (relative weight, 6.34). After these six factors, there was a significant drop in factor utilization. Each of the four remaining factors (percentage of education paid, computer skills, language skills and gender) received relative weight of 2.96 or lower by the recruiters. The large standard deviations across all objective weights obtained reveals that this sample of recruiters had broad differences in terms of the importance each attached to the decision factors included in this study. 43 TABLE 5. MEANS OF OBJECTIVE WEIGHTS BY DECISION FACTOR. STANDARD FACTOR MEAN' DEVIATION RANGE Career Objective 9.83 18.33 -0.1-77.2 Overall GPA 6.34 8.62 -0.6-35.7 Major GPA 9.91 13.34 -0.2-60.4 Work Experience 15.89 18.87 -0.3-81.0 Extra-Curricular Activity 16.00 18.94 -0.2-72.4 % of Education Paid 2.96 5.16 -0.4-23.5 Computer Skills 2.79 11.53 -0.7-82.5 Language Skills 2.72 4.65 -0.3-21.1 Willingness to Relocate 31.21 30.54 -0.1-87.3 Gender 2.21 3.16 -0.9-16.5 RECRUITER SUBJECTIVE DECISION STRATEGIES An alternative measure of recruiter decision strategies was derived from the subjective/ stated weights recruiters were asked to assign to the ten decision factors included in the studyu Table 6 presents the means of the recruiter subjective weights. These recruiters perceived that they rely most upon a candidate's work experience (subjective weight of 25.78) when making pre-screening decisions. Willingness to relocate, extracurricular activities and major grade point average (with weights of 13.90 and 10.14 subjective 15.14, 44 respectively) were the next most heavily weighted subjective decision factors. Recruiters stated that the next most common factors they considered were career objective (subjective weight of 9.10) , overall grade point average (subjective weight of 8.00) and percentage of education paid (subjective weight of 7.22). The factors of computer skills and language skills followed (subjective weights of 5.51 and 4.92 respectively), while gender was the lowest weighted decision factor (subjective weight of 0.29) recruiters. TABLE 6. for this sample of MEANS OF SUBJECTIVE WEIGHTS BY DECISION FACTOR. STANDARD FACTOR MEAN DEVIATION RANGE Career Objective 9.10 10.27 0-50 Overall GPA 8.00 4.91 0-20 Major GPA 10.14 6.45 0-30 Work Experience 25.78 11.93 10-50 Extra Curricular Activity 13.90 8.19 0-39 % of Education Paid 7.22 6.14 0-30 Computer Skills 5.51 5.29 0-20 Language Skills 4.92 4.25 0-15 Willingness to Re-locate 15.14 11.12 0-50 Gender 0.29 1.12 0-05 45 OBJECTIVE (STATISTICAL) WEIGHTS VS. SUBJECTIVE (SELF-REPORT) WEIGHTS The first and most significant objective of this study was to determine the relationship between recruiter actual and perceived decision factor usage when making decisions about which candidates to interview. This relationship will be explored by comparing the objective (statistical) weighting schemes obtained through the policy capturing analysis with the subjective (self-reported) weights provided by the recruiters. The relation between the two sets of weights was assessed through two different methods. First, individual-level Spearman rank order correlations were computed between the two sets of weights. Because the two sets of weights were on a common metric (where both the relative and subjective weights summed to 100) the correlations could be computed directly (Zededk and Kafry, 1977). Table 7 presents the frequency distribution for the Spearman rank order correlations obtained. The mean correlation was .553, with a range of values from -.031 to .855 (see Appendix B for individual values). As can be seen, there was relatively good agreement between the two sets of weights, indicating that most recruiters were aware of the policies that they used when making their judgements. Thirty- three of the 51 recruiters (64.7%) had correlations of .50 or 46 better, while only one recruiter had a correlation that was negative in sign. TABLE 7. FREQUENCY DISTRIBUTION OF‘THE SPEARMAN RANK-ORDER CORRELATION BETWEEN RELATIVE AND SUBJECTIVE WEIGHTS. I RANGE l FREQUENCY I -.19 - .00 1 .01 - .19 2 .20 - .29 5 .30 - .39 6 .40 - .49 4 .50 - .59 7 .60 - .69 7 .70 - .79 15 .80 - .89 4 .90 - .99 0 The statistical and subjective weights were also analyzed using t - tests to measure the mean difference for each of the weightings. Hypothesis 1 (e.g. there is significant difference between hospitality recruiter objective and subjective decision factor usage when making pre-screening decisions), was formulated to obtain more specific evidence regarding the study's first objective. This hypothesis was tested, using SPSS Subprogram T-Pairs, with the corresponding statistical null hypothesis: 47 H01: There is no significant difference between hospitality recruiter objective and subjective decision factor usage when making pre-screening decisions. This hypothesis was not retained at the .05 level. Significant differences were found for five of the ten decision factors considered. Table 8 presents the results of this analysis. There were significant differences (even at the more stringent alpha = .01) between the statistical and subjective weights for the factors of work experience (t= 4.11, p = .000), percentage of education paid (t = 4.83, p = .000), language skills (t = 3.26, p = .002), willingness to re-locate (t = -5.13, p = .000) and gender (t = -4.04, p = .000) . The statistical and.subjective‘weights for the factors of career objective (t = -0.43, p = .670), overall grade point average (t = 0.15, p = .162), major grade point average (t = 1.42, p = .879), extra—curricular activity (t = -0.93, p .= .356) , and computer skills (t = 3.26, p = .067) were not significantly different from one another (alpha = .05). Overall,_,the—results suggest thatalthough the pattern or shape of the distributions of the statistical and subjective weights were similar (as evidenced by the Spearman rank-order correlations), the magnitude of the factor weights were somewhat dissimilar (as evidenced by the t-test of mean differences). 48 TABLE 8. T-TEST BETWEEN THE OBJECTIVE AND SUBJECTIVE WEIGHTS BY FACTOR. OBJ. SUBJ. WEIGHT WEIGHT T SIGNF. FACTOR MEAN S - D . MEAN S - D - VALUE VALUE Career Obj. 9.83 18.33 9.10 10.27 -0.43 .670 Overall GPA 6.34 8.62 8.00 4.91 1.42 .162 Major GPA 9.91 13.34 10.14 6.45 0.15 .879 Work Exp. 15.89 18.87 25.78 11.93 4.11 .000 Extra Curric. 16.00 18.94 13.90 3.19 -0.93 .356 % Of Ed. 2.96 5.16 7.22 6.14 4.83 .000 Comp. Skills 2.79 11.53 5.51 5.29 1.88 .067 Lang. Skills 2.72 4.65 4.92 4.25 3.26 .002 Wlng.tO Reloct 31.21 30.54 15.14 11.12 -5.13 .000 Gender 2.21 3.16 0.29 1.12 -4.04 .000 RECRUITER RATING CONSISTENCY The multiple R2 Obtained from the policy capturing analysis allows for an examination Of the consistency with which recruiters made their ratings. The mean R2 for the sample was 0.718, with a range Of values from 0.473 to 0.885. Table 9 presents a frequency distribution Of R? for the 51 raters (Appendix B includes a table Of individual recruiter R2 values). Most recruiters were consistent in using their rating strategy across the 40 profiles. The few raters whose Ig'were lower, probably either did not understand or carefully consider the candidate profiles. 49 TABLE 9 . FREQUENCY DISTRIBUTION OF THE R2 VALUES OBTAINED FROM THE POLICY CAPTURING ANALYSIS. I RANGE l FREQUENCY I 0.00 - .09 0 .10 - .19 O .20 - .29 0 .30 - .39 O .40 - .49 2 .50 - .59 4 .60 - .69 10 .70 - .79 27 .80 - .89 8 .90 - .99 0 Recruiters answered a question in the HRQ concerning how consistent they felt they were in making their ratings. In answering the question: "To what extent did you use the same factors tO rate all applicants?", none Of the respondents selected answer #1, (to a very minimal extent). One recruiter (2.0%) selected answer #2, (to a small extent), while 6 recruiters (11.8%) chose answer #3 (to a moderate extent). Answer #4 (to a large extent) was reported on 37 (72.5%) of the HRQ's and answer #5 (to a very great extent) was selected by 7 recruiters (13.7%). This self-report information on 50 consistency relates favorably with the statistical findings concerning the R2 obtained. Table 10 presents the means and standard deviations for the R2 by response category. In general, higher R2 values are associated with self-report responses Of higher consistency. .An analysis Of variance was computed to further analyze the relationship between the R2 values and the self-reported consistency measure. The ANOVA revealed that the groups were significantly different from each other at the .05 level Of significance (F3,,.7 = 2.98, p = .041) when considering self-reported level Of consistency and mean R2 values. TABLE 10 RELATIONSHIP BETWEEN R2 VALUES AND SELF-REPORT MEASURE OF CONSISTENCY. Self Rating - In Response to the Question "TO what extent did you use the same factors R2 to rate all applicants?" n Mean S.D. (1). to a very minimal extent 0 - - (2). tO a small extent 1 .49 - (3). to a moderate extent 6 .72 .05 (4). to a large extent 37 .71 .10 (5). to a very great extent 7 .77 .08 DECISION FACTOR IMPORTANCE BY INDUSTRY SEGMENT The second Objective Of this study addressed the relationship between decision factor usage and the hospitality 51 industry segment recruiters are employed within. Table 11 presents the mean Objective and subjective decision weights and standard deviations for each factor classified by industry segment. Research hypothesis 2 (e.g. there is significant difference among recruiters from different segments Of the hospitality industry with respect to their Objective decision factor usage when making pre-screening decisions) was formulated to Obtain more specific evidence regarding the study's second Objective. This hypothesis was tested, using multivariate analysis of variance technique (SPSS Subprogram MANOVA), with the corresponding null hypothesis: sz: There is no significant difference among recruiters representing different hospitality industry segments with respect to their Objective decision factor usage when making pre-screening decisions. The null hypothesis was not retained at the .05 level. Significant difference was found among recruiters from different hospitality industry segments with respect to their mean Objective decision weights (F = 1.89, p = .015). .3OHOQ haoumwooaaw mcowumw>mo oumocoum I momma o>wu00nn5m .m .SOHOQ wawuoprESH mGOHDOH>mO unmocmum I momma O>Huomnno .H 8.6 an; 3.~ $3 8.6 ~m.~ 8.6 3.6 8.6 3.4 66.6 ~m.~ 3.3 3.3 66.6 mm; oo.o 3.6.. 66.6 38 .3980 3.3. 3.3” 3.3 3.3 8.: 3.3 3.3 3.3 3.6 8.3 Tom on 8.3 3.3 3.3 3.3 3.3 3.3 8.3 no.3 3.3 66.3 .633 $6 3.4 3.4 36 36 mm; maze 3.6 9.4 3.4 3386 3.6 3.... 36 case 3.~ 3.3 $3 3.6 $6 3.~ 6:3 3.4 m~.~ 66.4 3.6 -.m 3.3 3.3 3.2. 3.m AWN 333m oofi 3.~ 8;. no.6 3.3 3.3 co.» 3.3 36 an; .380 3K SJ. 36 SS 3.4 $6 3.3 mean 3.4 $3 053 3.6 mm.~ 3.» we... 8.6 $6 3.3 3.6 an... No.3 no w 35 3.3 $6 8.3 No.3 8.3 3.1. 3.6 moé 3.3 3.336 66.3 3.3 3.: 3.3 «6.3 3.3 mmfi 5.0 3.3 3.3 ouuxm 3.3 3.3 3.3 3.3 323 3.3. 3.3 3.3 3.3 3.3 .88 66.3 3.3 3.3 3.3 8.3 3.3. no.3 3.3 3.3 3.6 0783 5.4 3.3 3.4 omé 3:4 8.: 3.... om.o $6 3.3 «no 85 3.3 cod 3.6 3.6 36 no.3 3.6 3.3 3.3 mono: «66 3.4 oo.m 3.6 36 3.6 34... 36 21m 33 go 8.3 3.6 $6 3.6 oo.m 36 36 3.3 om.w 3.» .236 3.1” 3.6 3.3 3.3 3.6 3.3 666 SS 3.3 3.3 ..30 3.4 3.6 no.3 3.3 5.3 no.6 66.3 35m $6 $6 .8058 .m. d. m o o o m o ~m .o moeoam __ cameo .560 zoz 353.5 3:... noon soon 6333 _ .azmzomm smamoozH squaeHmmom um mozaemomzH moaoem .33 mamas 53 Univariate F-tests demonstrated that a statistically significant difference could be found on the rating of the factor "computer skills" (F‘_“ = 4.73, p = .003) while the factor "work experience" approached significance (F‘M = 2.46, p = .058). No other mean weights approached significance at the .05 level. The significant difference for the factor of computer skills was examined (at the .05 level) using Duncan's multiple range test (SPSS Subprogram DUNCAN). The Duncan's test demonstrated that fast food restaurant recruiters put significantly greater weight on computer skills (when making pre-screening' decisions) than. did. the four other' groups considered. Hypothesis 3 (e.g. there are significant differences among recruiters representing different hospitality industry segments with respect to their subjective decision factor usage when making pre-screening decisions) was also formulated to obtain more specific evidence regarding the study's second objective (see Chapter 1). This hypothesis was tested using multivariate analysis of variance technique (SPSS Subprogram MANOVA), with the corresponding null hypothesis: IgB: There is no significant difference among recruiters representing different hospitality industry segments with respect to their subjective decision factor usage when making pre-screening decisions. 54 This hypothesis was retained at the .05 level of significance. No significant differences were found among recruiters from different hospitality industry segments with respect to their mean subjective decision weights (F = 1.24, p = .167). However, the follow-up univariate F-tests did reveal significant differences on the factors of gender (3..“ = 2.93, p = .031) and percentage of education paid (F1A6== 2.58, p = .049). No other decision factors approached significance at the .05 level. Duncan's multiple range tests for these two decision factors revealed that recruiters from the non-commercial foodservice segment placed significantly greater weight on gender than did their colleagues from lodging , full service (commercial) restaurants or the "other" category (at the .05 level). Given that the recruiters were not asked to explain their subjective weighting of gender, it is not possible to ascertain whether they rate male or female candidates more favorably. The Duncan's test also revealed that the lodging recruiters placed significantly less weight on the factor of percentage of education paid when compared to their colleagues from full service restaurants and fast food restaurants (again at the .05 level of significance). 55 DECISION FACTOR IMPORTANCE BY INDIVIDUAL RECRUITER DIFFERENCES ; The third objective of this study addressed the relationship between recruiter decision weights and individual recruiter differences. Multivariate analysis of variance techniques were used to determine the impact of individual recruiter differences on the weighting of the objective and subjective decision factors. These statistical analyses considered.the ten mean objective decision weights and.the ten mean subjective decision weights to assess the relationship between them and the following individual recruiter differences (as measured by questions on the Hospitality Recruiter Questionnaire): 1.) number of times served as a recruiter on college campuses 2.) number of years served recruiting on college campuses 3.) number of years of hospitality industry operations experience 4.) number of hours of special training in recruiting received 5.) age 6.) gender 7.) level of involvement in college recruiting (full time or regular vs. not regularly involved/occasionally asked to help out) 8.) level of input in the hiring process Research hypothesis 4 (e.g. there is significant difference in hospitality industry recruiter objective decision factor usage with respect to individual recruiter differences) had been formulated to obtain more evidence 56 regarding the study's third objective. This hypothesis was tested, using SPSS Subprogram MANOVA, with the corresponding statistical null hypothesis: H¢4z There is no significant difference among hospitality industry recruiter objective decision factor usage with respect to individual recruiter differences. The null hypothesis was retained at the .05 level. No significant differences in objective decision factor usage were found with respect to any of the eight individual recruiter differences considered in this study. The univariate F-tests used to examine each of the variables separately also resulted in no significant differences found at the .05 level. As a result, no further statistical analyses of the relationship between recruiter objective decision factor usage and individual recruiter differences was appropriate. Research Hypothesis 5 (e.g. there is significant difference in hospitality recruiter subjective decision factor usage with respect to individual recruiter differences) was also formulated to obtain more evidence regarding the study's third objective (see Chapter 1). This hypothesis was tested, using SPSS Subprogram MANOVA, with the corresponding statistical null hypothesis: H05: There is no significant difference among hospitality industry recruiter subjective decision factor usage 57 with respect to individual recruiter differences. The null hypothesis was retained at the .05 level. No significant differences in subjective decision factor usage were found with respect to any of the eight individual recruiter differences considered in the study. The univariate F-tests used to examine each of the variables separately also resulted in IN) significant differences at the .05 level of significance. As a result, no further statistical analyses of the relationship between recruiter subjective decision factor and individual recruiter differences was appropriate. SUMMARY This chapter included a description of the sample, as well as a report of the findings derived from the research methods utilized. Chapter 5 will present a summary of the key findings from which conclusions and recommendations for further research will be drawn. V. S CONC U I NS D RECOMMEND TIONS INTRODUCTION The intent of this chapter is to provide a summary of the study, to draw conclusions for practice (especially for the major groups involved with or interested in hospitality recruiting) , and to make recommendations for further research. SUMMARY The primary purpose of this study was to examine how hospitality recruiters made decisions regarding who to interview. This was tested by examining recruiter pre- screening decision strategies to determine if their objective and subjective decision factor usage was different. This study also examined decision factor usage among the different segments of the hospitality industry. Lastly, this study examined hospitality recruiter decision factor usage with respect to individual recruiter differences. The Hospitality Recruiting Questionnaire (HRQ) was developed (based on an instrument first used by Gardner, Kozlowski and Hults in their 1988 study of general recruiters) and tested, and then administered to a sample of hospitality industry recruiters who attended the Michigan State 58 59 University, School of Hotel, Restaurant and Institutional Management's CAREER EXPO XIV on November 17, 1992 in East Lansing, MI. The 51 recruiters who completed the HRQ provided information regarding their age, gender, recruiting experience, recruiting training and hospitality industry operations experience. Participants further indicated their level of input in the hiring process as well as whether their organizations provided them with criteria to use when making pre-screening decisions. Recruiters also indicated which segment of the hospitality industry their organization competes within. Data relative to the recruiters' ratings of 40 employment candidate profiles in the HRQ were analyzed using T-tests of mean differences (SPSS Subprogram T-TEST) as well as multivariate analysis of variance (SPSS Subprogram MANOVA) and Duncan's multiple range test (SPSS Subprogram DUNCAN). The researcher found that for this sample of hospitality recruiters: 1. There were significant differences in the recruiter mean decision factor weights for five of the ten decision factors considered in the study. 2. When considered as a whole, there were significant differences among recruiters from different hospitality 60 segments with respect to their overall mean objective decision factor usage when making pre-screening decisions. When each factor was examined individually, recruiters from the fast food segment placed significantly greater weight on a candidate's computer skills than did the recruiters representing the other segments considered in this study. 3. When considered as a whole, there were no significant differences among recruiters from different hospitality industry segments with respect to their overall mean subjective decision factor usage when making pre- screening decisions. When each factor was examined individually, recruiters from the non-commercial segment placed significantly more weight on a candidate's gender than did their colleagues from lodging organizations, full service restaurants and the "other" category. Further, lodging recruiters placed significantly less weight on the percentage of education paid decision factor than did the recruiters from fast food and full service restaurant organizations. 4. The study revealed that individual recruiter differences had no significant impact upon either objective or subjective mean decision factor usage (when considered as a whole or by individual variable) by 61 recruiters when making pre-screening decisions. The :researcher' also found. relatively large standard deviations for most of the decision factors included in the candidate profiles. Large standard deviations indicate a wide variation in factor usage by the recruiters in this sample. The multiple R2 obtained from the policy capturing analysis was used as evidence of rater consistency. The high E? values obtained in this study indicate that recruiters were employing a consistent strategy in making the ratings across the profiles. Additionally, recruiters with higher R2 values were more likely to self-report that they were consistent in making their ratings. The Spearman rank order correlation analysis of recruiter objective and subjective decision factor usage revealed that the pattern of the weighting of the cues between the two sets of weights were fairly similar; Most.recruiters seemed to have a good awareness of the ranking of the importance with which the decision factors were used in making their ratings. When the relationship between the objective and subjective weights was examined in terms of the magnitude of the weights placed on each, recruiters seemed less aware of the weights used. It seems the participants can estimate the rank ordering of the decision factors that they use in assigning ratings but are less accurate in estimating the statistical (actual) weights used. 62 CONCLUSIONS The findings of this study provide a significant amount of thought. provoking information for ‘thre xtmajor' groups involved in or concerned with hospitality recruiting: hospitality recruiters, hospitality students and hospitality educators. For this sample of hospitality recruiters, this study revealed some interesting findings. As a group, these recruiters 'were relatively’ consistent in 'using' the same factors to evaluate candidates for interviews and they were relatively aware of the ranking of decision factor importance as they made pre-screening decisions. However, of the ten decision factors recruiters used in the study, only three of these factors (e.g. willingness to re-locate, extra-curricular activity and work experience) accounted for over 60% of the recruiters' actual decision strategies. Willingness to re-locate (or the lack thereof) accounted for approximately 31% of the decision strategy, twice as much as any other decision factor included in the study. This finding was consistent with earlier studies of this sort. Further, while recruiters were relatively aware of the ranking of the weights of their decision factors, five of the ten factor mean weights in this study were significantly 63 different with respect to the recruiters perceived weighting vs. his/her actual weighting. Recruiters tended. to overestimate the importance of work experience when comparing actual to perceived decision weighting and to underestimate the importance of a candidate's willingness to re-locate. Additionally, in the case of both language skills and percentage of education paid, recruiters believed they weighted these factors more heavily than they actually did when rating the candidate profiles. When considering the gender of a candidate, this sample was statistically significant in terms of the difference between their actual and perceived weighting of this factor. Given this statistical difference, further analysis was undertaken to determine the practical significance (if any) of this difference. A review of the sign attached to the standardized beta weights (generated during the policy capturing analysis) was used to assess the recruiters preference in terms of male versus female candidates. Because male candidates were coded "zero" and female candidates were coded "one", a negative sign attached to the beta weight for gender indicated males did better with this recruiter, whereas a positive beta indicated females were the favored gender; 0f the 51 recruiters in the sample, 30 (58.8%) had negative beta weights for the gender factor (i.e. they favored male candidates) while the remaining 21 (41.2%) had positive beta weights for the factor of gender (i.e. they favored female 64 candidates). Broken.out.bngender as illustrated.in Table 12, 21 of the male recruiters favored male candidates and the remaining 14 male recruiters favored female candidates. Nine female recruiters favored males, while the remaining seven female recruiters favored females. In considering the practical significance of these analyses, it is important to note that as a group, these recruiters believed they placed virtually no weight on gender as a decision factor. However, the actual weighting of gender demonstrated otherwise. While the majority of the recruiters placed very little to no actual (statistical) weight on gender, the decision weight standard deviations and a review of the individual recruiter decision strategies (see.Appendix:B) reveals that one of the recruiters in this sample was measured as having a relative weight as high as 16.5 and several others displayed relative weights greater than 4.0 in their weighting of gender when rating the candidate profiles. Given the laws of the United States and general societal concerns about gender discrimination, this finding may serve as a "red flag" for those recruiters who (consciously or otherwise) are using gender when making pre- screening or any other recruiting related decisions. 65 TABLE 12. RECRUITER BETA WEIGHT SIGNS FOR THE FACTOR OF CANDIDATE GENDER. POSITIVE BETA WEIGHT FOR GENDER (i.e. FAVORED MALES) NEGATIVE BETA WEIGHT FOR GENDER (i.e. FAVORED FEMALES) RECRUITER GENDER Male (35) 21 14 Female(16) 9 7 TOTAL (51) 30 21 In analyzing the overall subjective decision strategies of this sample (what the recruiters "thought" they did) it was not surprising to find no statistically significant differences. However, though. they' represent. a :relatively homogeneous group (i.e. hospitality industry only), there were statistically significant differences in the way they actually used the decision factors included in this study. Additionally, the standard deviations for both the objective and subjective decision weights were relatively large. As noted earlier, this was indicative of a lack of agreement among the recruiters in this sample as to which decision factors count most. When analyzing the factors individually, the industry segments differed significantly in their'weighting of only one of the objectiveadecision factors. Fast food recruiters placed significantly more objective weight on computer skills than 66 did the recruiters from other segments. In fact, the fast food recruiters were the only group in this sample to place much statistical (objective) weight on computer skills at all. Perhaps this is because of the nature of the fast food business (i.e. no guest check and strong sense of urgency in delivery of the product) which necessitates tighter financial controls due to very tight profit margins. The benefits of computerized_ systems 'may' be :more apparent to fast food operators, therefore the recruiters working for this segment may place more weight on an applicant's computer skills. Further discussion of computer skills and the apparent lack.of recruiter interest in them follows. In analyzing the recruiters' subjective decision factor use individually by industry segment, hotel recruiters‘did.not place as much weight on the amount of education personally financed by the candidate. Given the increasing costs of higher education, perhaps this sample of hotel recruiters see this factor as less significant.because more students may find it necessary to contribute to the financing of their degrees. Gender also showed up as receiving significantly different subjective weight from the non-commercial foodservice recruiters as compared to the other segments. However, a comparison of the subjective and objective weights of gender for non-commercial recruiters showed no significant difference between the two. While the non-commercial recruiters were different from their colleagues in that they 67 reported placing some (albeit small) weight on gender, they were also consistent and aware of this weighting as evidenced by the lack of significant difference between their stated and statistical mean weights. Individual recruiter differences such as age, gender and years of recruiting and/or hospitality experience did not have an impact on how these recruiters decided which candidates to interview. While it may have been expected that more experienced or better trained recruiters would have been different from their less experienced or less trained colleagues, findings indicated no such statistical relationship for this sample. Though the review of literature indicated the current and anticipated importance of computer skills and language skills for success as a hospitality manager, recruiters in this sample did not place much weight on either of these factors. Recruiter subjective and objective weighting of these factors were not significantly different. Perhaps these factors come into play later in. the recruiting' process; but neither computer skills or language skills was perceived to be or actually shown to be very important to this sample of recruiters. For hospitality students, it is important to note that recruiters in this samplejplaced.the:greatest.objective factor weighting on a candidate's willingness to re-locate. Students may want to think carefully about how and when to reveal 68 information regarding their willingness to re-locate. Findings of this study indicate that if a student were to include information which suggests a restriction or unwillingness to relocate on their resume, the student.risks the possibility of reducing or even eliminating their opportunities to interview. 0n.the other hand, indication of willingness to move where the employer has needs, could potentially work in favor of a student seeking the opportunity to interview. Students should also be aware that hospitality recruiters in this sample weighted work experience and involvement in extra-curricular activities in a virtual tie as the second most highly weighted decision factors. While academic performance (both major and overall grade point average) counts when it comes to pre—screening decisions, these findings indicate that they are not as important as performance outside the classroom. Students with interests in specific segments within the hospitality industry should be aware of the differences in decision factor usage:by segment.noted earlier. However, given the relatively small sample sizes of each segment included (especially fast food, n = 3) caution should be taken in interpreting these results. Two points are important for students to keep in mind when considering the results of this studyu The first is that while the sample includes recruiters from some of the largest hospitality organizations in the world and is arguably 69 representative of the major employers of MSU-HRIM graduates, it was not randomly drawn and was relatively small overall (n = 51). This becomes even more obvious when the sample is segmented” 'The second.point.is that large standard deviations in recruiter decision weights are indicative of the many different opinions recruiters have about what is important when deciding which candidates to interview. This second point also presents students with a strategy to assist in surviving the pre-screening process. In addition to researching annual reports, recruiting brochures, corporate mission statements and the like, it may be quite useful for students to discover more about each recruiter individually. Through opportunities such.as conversations at‘career fairs or during classroom visits/ guest lectures, or through the advice of faculty who are well acquainted with given recruiters, it is possible for a student to find out what really matters to the specific individual who will be deciding whether to interview them - or not. Armed with this knowledge, the student may proceed appropriately. For hospitality administrators and faculty, this study also provides valuable information. For example, if students are to utilize faculty as a resource to learn about recruiters, faculty might consider balancing their research and teaching responsibilities with opportunities to network with recruiters. The findings of this study also point to the issue of the 70 relative importance of grades. Recruiters in this sample did not weight consideration of grade point average (either objectively or subjectively) as among the most important factors when making pre-screening decisions. It is possible this means that grades are not very important when pre- screening decisions are made. It may also be true that recruiters expect any student accepted into a major university program (such as the School of Hotel, Restaurant, and Institutional Management at Michigan State University, which was the site of this study) will have sufficient intelligence to succeed, therefore grades do not effectively differentiate candidates relative to pre-screening decisions. And yet at the time of this study, the School of Hotel, Restaurant, and Institutional Management.at.Michigan State University limited access to its program to only those students who receive 2.5 or better (4.0 scale) in six courses outside of the hospitality major curriculum (i.e. accounting, economics, math, computer science). If, in fact, grades are not important to recruiters, the findings of this study may point out the possibility that hospitality academia may, in some cases, over-emphasize the importance of grade point average to the detriment of some of the very individuals it purports to serve. Most hospitality management programs have moved to include (and in some cases to emphasize) the use of computers and the development of additional language skills in their 71 curriculum. As noted earlier, recruiters in this sample did not place much weight on either the computer skills or language skills.decision factors included.in the studyu These results would seem to indicate an over emphasis in this area. A final observation regarding the results of this study is that the recognition of absolute requirements for success as an employment candidate in the hospitality workplace are not yet available, or expected. In that sense, this research might indicate that there are simply too many variations (industry segments etc.) to recommend specific courses of action to all students or recruiters or faculty; 0n the other hand, an alternative explanation may be that these skills are considered later in the hiring process and.that they simply do not merit much weight at the pre-screening stage. It is also possible, especially in the case of computer skills, that the employers expect. to train. their new-hires on their own specific computer system and therefor do not put much weight on this factor as a pre-screening criteria. RECOMMENDATIONS FOR FURTHER RESEARCH The following recommendations are suggested by the researcher based upon the findings and conclusions of this study: 1. These results should be viewed as preliminary. Given the sampling method employed and the relatively small sample 72 size obtained, a replication of this study with a larger, randomly drawn sample should be attempted to determine if the same results are identified. 2. This study examined recruiting only from the perspective of recruiters. A modification of the HRQ could be employed to study the perceptions of hospitality students as well as hospitality faculty, relative to the decision weights they each believe recruiters utilize when making pre-screening decisions. A study of this sort would provide an opportunity to compare the perceptions of the three groups most concerned with hospitality recruiting. 3. The results of the policy capturing analysis used in this study were useful in determining the relative weights of the decision factors recruiters considered when making pre- screening decisions. Conducting tests of statistical significance between the recruiter objective and subjective mean decision weights was useful in further analyzing how recruiters decide who to interview. However, policy capturing is limited in the level of detail it provides relative to how the rater uses or combines decision factors (Einhorn, Kleinmuntz, & Kleinmuntz, 1979). Future research could employ verbal and/or written protocols as a supplement to the policy capturing analysis. Such rich and detailed information as the order in which hospitality recruiters look at cues and the number of cues looked at can be examined with this approach. 4. This study examined only one (e.g. pre-screening 73 decisions) of a series of steps in the recruiting process. Future research could take a :more holistic approach to investigate how candidate information is transferred throughout the entire process. This may be best accomplished through ethnographic methodologies, possibly involving such strategies as the interviewing of recruiters and candidates, audio and/or video taping of employment interviews and document analysis (i.e recruiter interview notes, correspondence between. candidates and. recruiters and recruiters and actual business units etc.). BIBLIOGRAPHY Anderson, B. (1977). " Differences in teachers' judgement policies for varying numbers of verbal and numerical cues." Or anizational Behavior and Human Performance, 19: 68-88. Balzer, W., Rohrbaugh, J., and Murphy, K. (1983) . "Reliability of actual and predicted judgements across time." Organizational Behavior and Human Performance, 32: 109- 123. Billings, R., and Marcus, 8. (1983). "Measures.of compensatory and non-compensatory models of decision behavior: Process ‘tracing ‘versus policy capturing." Organizational Behavior and Human Performance, 31: 331- 352. Borg, W., and Gall, M. (1989). Educational Research (5th edition). New York, NY: Longman Publishing. Brady, D., and.Rappaport, L. (1973). Policy-capturing in the field: The nuclear safeguards problem. nggpiggpigpg; Behavior and Human Performance, 9: 253-266. Bray, J. and Maxwell, S. (1985). HEW—g: Variance. Sage University Paper series on Quantitative Applications in the Social Sciences, 07-054. Beverly Hills: Sage Pubns. Campion, M. A. (1987). “Identification of variables most influential in determining interviewers' evaluations of 74 75 applicants in the college placement center." Psychological peports. 42: 947-52. Cattin, P. (1980)."Estimation of the predictive power of a regression model." Journal of Applied Psychology, 65: 407-414. Cook, R., and Stewart, T. (1970). " A comparison of seven methods for obtaining subjective descriptions of judgmental policy." Organizational Behavior and Human Performance, 13: 31-45. Dawes, R. (1979). "The robust beauty of improper linear'models in decision making." American Ps cholo ists, 34: 571- 582. Dawes, R.M., and Corrigan, B. (1974) "Linear models in decision making." Psychological Bulletin, 81: 95-106. Doherty, M., and Keley, S. (1972). "Use of subjective predictors in regression analysis for policy capturing." Journal of Applied Psychology, 56: 277-278. Durocher, J. F. and Goodman, R. J. (1991). "Training Program Expectations: A Conundrum." The Cornell HRA Quarterly, 2: 76-78. Ebert, R., and Kruse, T. (1978). "Bootstrapping the security analyst." Journal of Applied Psychology, 63: 110-119. Einhorn, H. (1971). "Use of nonlinear noncompensatory models as a function of task and amount of information." Organizational Behavior and Human Performance, 6: 1-27. Einhorn, H., Kleinmuntz, D., and Kleinmuntz, B. (1979). 76 "Linear regression and process-tracing models of judgment." Psychologigal Review, 86: 465-485. Farmer, S. W. and Tucker, D. (1989) "A comparative study of working conditions in the hospitality industry. Expectations versus realities." Hospitality Educatiop gnd Researgh Journal. 2: 12-26. Feldman, M. L. andfiMurata, D.K. (1991). "Why Mergers Often Go ‘Pfft!'" ABA Banking Journal. 8:34-36. Ford, J.K., Schmitt, N., Schechtman, S., Hults, B. and Doherty, M. (1989). "Process Tracing Methods: Contributions, Problems and Neglected Research Questions." nggnizational Behaviop and Human Qecisigp Progesses, 43:, 75-117. Gardner, P.D., Kozlowski, S. W. and Broadus, A. (1988). "Egg Ideal Candidate: What Recruiters Would like to See." Collegiate Employment Research Institute. East Lansing, MSU. 8 p. Gardner, P.D, Kozlowski, S. W. and Hults, B. M. (1988). Screening Job Applicants: Recruiter Decision Strategies. Collegiate Employment Research Institute. East Lansing, MI. 26 p. Glass, Gene and Stanley, Julian. (1970). Statistical Methods in Education and Psychology. Englewood Cliffs, NJ.: Prentice-Hall. Goldberg, L. (1971). "Five models of clinical judgment: An empirical comparison between linear and nonlinear V 77 representations of the human inference process." Organizational Behavior and Human Performance, 6: 458- 479. Hanigan, M. (1991) . "Seizing the Competitive Edge on Campus." Managers Magazine. 8:15-17. Harcourt, J. and Krizam, A. C. (1989). "A Comparison of Resume Content Preferences of Fortune 500 Personnel Administrators and Business Communication Instructors." Journal of Business Communication. 2: 177-190. Hedderson, J. (1987). SPSS-X Made Simple. Belmont, CA: Wadsworth Publishing Hiring Costs and Stpategies: The AMA Report. (1986). American Management Association. New York: AMACOM Briefings. Hobson, C.J., and Gibson, F.W. (1983). "Policy capturing as an approach to understanding and improving performance appraisal: A review of the literature." Academy of Management Revigy, 8: 640-649. Hoffman, P.J. (1960). "The paramorphic representation of clinical judgment." Psychological Bulletin. 57: 116- 131. Holley, W. J. Jr., Higgins, H. and Speights. (1988). "Resumes and Cover Letters: What Do HR Managers Really Want?" Personnel. 12: 49-51. Huber, V. L., Northcraft, G. B., and Neale, M.A. (1990). "Effects of Decision Strategy and Number of Openings on 78 Employment Selection Decisions. " Organizational Behavior d Hu a s o c sse . 2: 276-284. Klaas, B. S. and Dell 'Omo, G. G. (1991). "The Determinants of Disciplinary Decisions. The Case of Employee Drug Use." Persoppel Psychology. 4: 813-835. Knutson, B. J. (1989). "Expectations of hospitality juniors and seniors: Wave II." Hgspitality Education ppg Research Journal. 3: 193-201. Knutson, B. J. (1989). "Hospitality alumni survey: Were their expectations as graduates met as industry employees?" Hospitality Education and Research Journal. 3: 463-468. Knutson, B. J. (1987). "Hospitality student survey: Measures of influences on and expectations for post- graduate employment" Hospitality Education and Research Journal. 2: 163-174. Knutson, B. J. and Patton, M. E. (1992). "How Prepared.Am I to Succeed in the Hospitality Industry?" Hospitality and Tourism Educator. 3: 38-44. Laker, D. and.Gregory S. (1989). "Attraction and Influence on the Employment Decision of the Recent Hotel and Restaurant Grad." Hospitality Education and Research Journal. 3: 31-42. Ley, D. A. and Sandler, M. (1982). "Corporate Recruiters: What Do They Want? Cornell HRA Quarterly. 2: 42-45. Martin, S. L. and Raju, N. S. (1992). "Determining Cutoff 79 Scores That Optimize Utility: A Recognition of Recruiting Costs." Journal of Applied Psychology. 1: 15-23. McCleary, K. W. and Weaver, P. A. (1988). "Expectations of hospitality students regarding entry level positions in the hospitality industry." Hospitality Education and Research Journal. 2: 163-174. Michigan State University Placement Services, Follow-up Reports (1985-1991). East Lansing, MI. Miller, G.A. (1956). "The magical number seven, plus or minus two: Some limits on our capacity to process information." Psychological Review, 63: 81-97. Muchinsky, P. and.Harris, S. (1977). "The effect of applicant sex and scholastic standing on the evaluation of job applicant resumes in sex-typed occupations." Journa of Vocational Behavior. 11: 95-108. Nisbett, R.E., Krantz, D. H., Jepson, C., and Kunda, Z. (1983). "The use of statistical heuristics in everyday inductive reasoning." Psychological Review. 90: 339- 363. Nisbett, R. E. and Wilson, T. D. (1977). "Telling more than we know: Verbal reports on mental processes." Psychological Review. 84: 231-259. Ogilvie, J., and Schmitt, N. (1979). Situational influences on linear and nonlinear use of information. Organizational Behavior and Human Performance, 23, 292- 80 306. Oskamp, S. (1967). "Clinical judgement for the MMPI: Simple or complex?" Journal of Clinigal Psychology, 23: 411-415. Payne, J., Braunstein, M., and Carroll,J. (1978). "Exploring predecisional behavior: An alternative approach to decision research." Organizational Rehavior ang Human Performance, 22; 17-44. Posner, B. Z. (1981). "Comparing recruiter, student, and faculty perceptions of important applicant job characteristics." Personnel Psycholggy. 34: 329-339. Reilly, B., and Doherty, M. (1989). A note on the Assessment of Self-Insight in Judgement Research. Organizational Behavior and Human Decision Procegggg, 44: 123-131. Rosseau, D. M. and Anton, R. J. (1991). "Fairness and Implied Contract Obligations in Job Terminations: The Role of Contributions, Promises, and Performance." Journal of Qrganizational Behavior. 4: 287-299. Scheetz, L. P. (1991). Recruiting Trends 1991-92. Placement Services. East Lansing: Michigan State University. Schenck, E. A., and Naylor, J.C. (1968). "A cautionary note concerning the use of regression analysis for capturing the strategies of people." Educational ang Psychological Measurment, 28: 3-7. Schmitt, N., and Levine, R. (1977). Statistical and subjective weights. Organizational Behavior and Human Performance, 20: 15-30. 81 Slovic, P., and. Lichtenstein, S. (1971). "Comparison. of Bayesian. and :regression approaches to ‘the study of information. processing in judgement." Organizatignal Behavior and Ruman Rerfgrnance, 6: 649-744. Slovic, Pl, Fischhoff, B., and. Lichtenstein, S. (1977). "Behavioral decision theory." Annual Review of Rsychology, 28, 1-39. Stahl, M. J. and Zimmer, T. W. (1984). "Modeling strategic acquisition policies: A simulation of executives' acquisition decisions." Academy of Management Journal, 27: 369-383. Svenson, O. (1979). "Process. descriptions of decision- making. "Or anizational Behavior and Human Performa ce, 23: 86-112. York, K. M. (1989) . "Defining Sexual Harassment in Workplaces: A Policy Capturing Approach." Academy of Manaqenent Journal, 4: 830-850. Zedeck, S., and Cascio, W. (1982). "Performance appraisal decisions as a function of rater training and purpose of appraisal." Journal of Applied Psychology, 67: 752-758. Zedeck, S., and Kafry,D. (1977). "Capturing rater policies for processing evaluation data. " WW and Human Performance, 18: 269-294. APPENDIX A 83 Zedeck and Cascio (1982) Algorithm GEN WTR LS CS PEP EA OGPA MGPA WE CASE CO 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 84 29 30 31 32 33 34 35 36 37 38 39 4O MICHIGAN STATE 35 U N l v E R s l T Y September, 1992 Dear Hospitality Recruiter: In conjunction with EXPO this year, we invite and encourage your voluntary participation in a unique research project. The goals of this research include identifying the factors that you (the hospitality recruiter) consider when reviewing candidates for employment and an analysis of how you decide which candidates to interview. Results of this study will be made available to each participating recruiter to provide you further insight into the thinking of your colleagues at other hospitath organizations. Findings of this study will also be useful to students as they prepare to embark upon their career journeys in the hospitality industry. For administrators and faculty of hospitality management schools, this study may hold implications for curriculum development and career counseling. All results will be reported in the aggregate, maintaining strict confidentiality of individual responses. Based on the significant number of people who stand to benefit from this study, we are confident that your participation will prove to be a good (and relatively short) investment of your valuable time. In an effort to make your participation as "painless and hassle-free" as possible, we will make the survey instrument and related research materials available at the Kellogg Center during CAREER EXPO XIV registration. Return of the completed study materials is easy. You can either drop them at the Kellogg Center front desk or at the Placement Center main desk (whichever is most convenient). The ultimate value of this project will be strongly influenced by the rate of participation of individuals like you. Please plan now to contribute your expertise and perspective by participating in this study during your visit to our campus. Should you have any questions about the study, please feel free to contact Mike Sciarini at (517)353-9211. Thank you again for your commitment to participate in CAREER EXPO XIV. We look forward to, seeing you in November! Yours in Hospitality, , nal F. Cichy, Ph.D. Director 86 NOTE TO THE READER: Data from items 13,14, and 16 is not reported in this dissertation. MSU HRIM MICHIGAN STATE UNIVERSITY SCHOOL of HOTEL, RESTAURANT and INSTITUTIONAL MANAGEMENT STUDENT and INDUSTRY RESOURCE CENTER HOSPITALITY RECRUITING QUESTIONNAIRE Thank you for agreeing to participate in this study that examines the decision strategies of hospitality recruiters. Before you begin, we want to remind you that the information used in this study is for research purposes only and that o r individ n will m tel nfi i . You indicate your voluntary agreement to participate by completing and returning this questionnaire. The study is divided into three parts: (1) background information about you and your company, (2) an evaluation of the suitability of a set of potential applicants, and (3) some questions about your reactions to the evaluation process. in order to facilitate your participation in this project, please complete each pan before examining the materials or moving on to the next part. When you have completed the survey, please place it in the return envelope and dmp it at the Kellog Center front desk 9: at the Career Development and Placement Services main desk 9: mail it by November 30, 1992 (we pay the postage). lfyou are interested in receiving a copy of the results of this study, please provide your name and address (or attach your business card) on the last page of this questionnaire Additionally, if you complete and return this questionnaire by November 30, 1992, you will become eligible for a drawing to award an array of prizes including an MSU-HRIM wristwatch, an MSU-HRIM sweater, and a complimentary booth registration for your company at the MSU-HRIM CAREER EXPO XV. The drawing will be held on December 1, 1992, and the winners will be notified by mail. lfyou haveany questions aboutthisstudy, please feel freetomntact MikeSciariniat (517) 353-9211. ‘lhankyou for your cooperation Please go on to the directions for Part 1. MM 1311'ch bum Whmofhmwm 1mm ranrr-aacxcnouun 87 mm ‘l‘he following questions ask about your recruiting experiences (section A), and your views on what factors Ire important in recruiting decisions (section 8). SECTION A - RECRUITIM EXPERIENCES AND EMPLOYER CHARACTERISTICS: 1. How many times have you served as a recruiter on college campuses? (use 0 for none) 2. How many years have you been a recruiter on college campuses? (use 0 for none) 3. How many years of hospitality industry operations experience do you have? (use 0 for none) 4. How many hours of special training or instruction in recruiting and/or interviewing have you had? (no for m) 5. What is your age? 6. Please circle your gender male female 7. What is your job title? 8. Which of the following best describes your involvement in college recruiting? _ College recruiting is a full-time or regular part of my job responsibilities __ i am occasionally asked to assist in college recruiting but u is not one of my regular job responsibilities 9. Which of the following categories m describes your organization? MammuWu-mmmm) _ ultimate! company _ Fast food restaurant company _ Full service restaurant company __ ConuaCVnon-commerthl food service company _ Other (please describe ) 10. Which of the following statements best describes your organization? independent/single unit - not aii‘iliated with a chain Mum-unit organization or chain 11. lfyou answered ‘multiunit organization or chain" for the prelious question, how many total unlts/ properties does your organization include? units/prepares 88 12. Does your company/organization provide you with specific criteria that you are to use when deciding which applicants to interview? YES NO 13. if you were reviewing the application/resume of a highly qualified individual but he/she indicated a regional preference outside of the area(s) in which you had positions available, what would you do? __ pass their credentials on to an appropriate colleague __ interview them anyway _ reject as not meriting an interview __ take some other action (explain below) 14. if you were reviewing the application/realm of a highly qualified individual but ire/she indicated a career objective not congruent with your organization, what would you do? _ pass their credentials on to an appropriate colleague _ interview them anyway _ reject as not meriting an interview __ take some other action (explain below) 15. Using the scale below, how would you rate your level of input in the hiring process? 1 2 3 4 5 pre-screen minimal moderate significant full hiring only input input input authority 89 SECTION B - RECRUITING DECISION FACTORS: 16. Please consider the factors that moould gram like to consider when reviewing HRIM student candidates for possible interviews. Under the “ideal Factors” column below, list all of these factors. There are no right or wrong arm's, we simply want you to indicate as accurately as possible your experienced opinion. Rating Ideal Factors Rating Ideal Factors Now please go back and evaluate the importance of each ideal factor, using the scale below. Place your evaluation in the “Rating” column to the left of the ideal factor. You may believe all the factors are important or that some factors are more important than others. Again, we are interested in your Opinion. r T ' T ' 1 2 3 4 S slightly somewhat moderately very extremely important important important important important to me to me to me to me to me PART 2 - APPLICANT EVALUATION This section contains a set of potential candidates for job interviews that you are asked to evaluate. Background about each candidate is provided in brief profile form (a mare complete description of the profiles follows). Please examine each profile and indicate the suitability of the applicant for an interview, using the rating scale below. Please record your rating on the “Rating" line to the right of the name of the individual you are evaluating For your convenience, this scale is repeated at the top of each page of profiles. We want your opinion about the applicants as if you were actually deciding which ones to interview. Please examineandratetheapplicantsoneatatime. Whenyou haveratedall applicants,goont0part3. T I I I j 1 2 3 4 5 unsuitable, moderately exceptionally not worth an suitable, suitable, interview might consider definite an interview interview 90 PLEASE READ THIS BEFORE MAKING ANY EVALUATIONS Tire profiles which follow have been developed based upon a review of each candidate’s credentials (including resume, Placement Services registration form and undergraduate transcripts). The information below is intended to assist you in understandingtheprofileformat. Forpurposesofthis study, thecandidatedatahavebeencondensed intorangesfor manyoftheaategoriesincludedineachproflle. Feelfreetoreferbacktothisinformationanecessaryduringthe evaluation process. EXPLANATION «CRITERIA: ° AlicandidatesareMSUHRlM majors, planningtograduateinMay,i993. - Winfomudonforeachcandidatewillbepmvided basedonthefollowinglevels: 1.) NONE PROVIDED.- candidatesinthisgroupdldnot lncludeanobjective. 2.) BROAD... resumesofthesecandidatessaythingssuch as‘toobtalnanentryleveihospitalitymanagement position" or similiar general statements. 3.) SPECIFIC... for purposes ofthls research, Interpret this level to refer to candidates whose objectives match with positions ypyr organization has available. ° W data for overall cumulativeGPA and GPA for major courses (using 4.0 scale) will be provided foreachcandidate Forpurposesofthis research, this datawill beprovidedbnedonthefollowingranges: 1.) between 2.0 and 2.6 2.) between 2.7 and 3.3 3.) between 3.4 and 4.0 0 W information wfll be provided utilizing the following descriptions: 1.) LEVEL I experience = a minimum of 400 hours of line level (hourly) hospitality industry work. Examples include (but are not limited to) guest service agent, housekeeping room attendant, host/hostess, bartender, server, dishwasher, cook etc. 2.) LEVEL 11 experience = a minimum of 400 hours of supervisory/management level hospitality work which may include rotation thnr multiple departments and/or completion of special projects. Examples include (but are not limited to) management trainee, management Intern, assistant manager, supervisor etc. NOTE: For purposes of the study, consider the work experience for each candidate as relevant to the type of positions you are hoping to fill. For example, if you recruit for hotels, interpret the work ocperience for each candidate as having occurred in hotels or other settings you may consider appropriate. ' WW will be categorized based on the following levels of participation: 1.) NONE.- if the candidate’s resume was void of activities. 2) MEMBER/PARTICIPAN'I'... if the candidate’ 5 resume mentioned participation or membership' rn one or more clubs or activities 3.) OFFICER/LEADER... if the candidate’ s resume noted one or more clubs or activities in which he or she was an officer and/or leader. 0 W refers to the amount oftotal college expenses the candidate was personally responsible for and will be expressed according to these ranges: 1.) 096-3396 2.) 34%-66% 3.) 6796-10096 91 - W will be described according to the following: 1.) LIMITED-. ifthecandldate’s GPAin computercourses bbetween 2.0 and 2.6 and hisor heresumeprovides no reference to experience with computers. 2.) AVERAGE- if the candidate’s GPA in computer courses is between 2.7 and 3.3 and his or her resume mentions limited experience with computers in hospitality work settings. 3.) ABOVE AVERAGE... if the candidate's GPA in computer courses is 3.4 or higher and his or her resume lists extensive on-the-job experience with spreadsheet and database management software. ' WWI detail the candidate’s language capabilities as follows: 1.) ENGLISH ONLY. 2.) ENGLISH PLUS (DNVERSANT... meaning the candidate is able to communicate on a basic level in one additional language beyond'English (assume the additional langrage would be useful in the worlquace) 3.) ENGLISH PLUS TLUEN’T... meaning the candidate is able to communicate on an advanced/sophisticated level in one language in addition to English (again where the additional language would be useful in the woriquace). ' W WI" be expressed 1* IOI'WSI l.) VERY LIMITED... the candidate has expressed a specific city or state they would prefer to live/work within but would consider alternatives if necessary. 2.) MODERATELY LIMITED... the candidate has expressed a multl-state regional preference but again would consider alternatives if necessary. 3.) UNLIMITED... the candidate is basically willing to go where the work is. 1 2 3 4 S unsuitable, moderately exceptionally not worth an suitable, suitable, Interview might consider definite an interview interview NEAL ADAMS: Rating CAREER OBJECTIVE Broad OVERALL GPA 2.7 - 3.3 MAJOR GPA 2.7 - 3.3 WORK EXPERIENCE One LEVEL I and Two LEVEL II EXTRACURRIC ACTIVITY Member/participant 96 OF EDUCATION PAID 34% - 66% COMPUTER SKILLS Average LANGUAGE SKILLS English plus conversant mattressro moms Moderately limited 92 T l 2 . unsuitable, not worth an interview LARRY PNIIIJPS: Rating __ CAREER oerrcnva suitable, might consider an interview OVERALL GPA MAJOR on WORK memos EXTRACURRIC. ACTIVITY 96 OF EDUCATION PAID COMPUTER SELLS exceptionally suitable, definite interview Specific 2.7 - 3.3 2.0 - 2.6 One LEVEL I and Two LEVEL 11 Officer/leader 0% - 33% Average LANGUAGE SELLS WILLINGNESS TO RELOCATE DOROTHY WOODS: Rating _ CAREER OBJECTIVE OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVITY % OF EDUCATION PAID COMPUTER SELLS LANGUAGE SKILLS WILLINGNESS TO RELOCATE GEORGE EVANS: Rating _ crease outcome OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVITY % OF EDUCATION PAID COMPUTER SELLS English plus fluent Moderately limited Broad 3.4 - 4.0 2.0 - 2.6 One LEVEL I and Three LEVEL II Member/participant 34% - 66% Limited English plus conversant Unlimited Specific 3.4 - 4.0 3.4 - 4.0 One LEVEL I and Three LEVEL 11 Officer/leader 67% - 100% LANGUAGE SELLS WELINGNESS TO RELOCATE Above average English plus fluent Unlimited 93 r— I 1 2 unsuitable, not worth an interview IANE'ISHEPHERD: Rating __ CAREER OBJECTIVE I 3 suitable, might consider an interview OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVITY % OF EDUCATION PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE USAFIIEY: Rating— CAREER OBJECTIVE OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVITY % OI" EDUCATTON PAID COMPUTER SILLS LANGUAGE SELLS WILLINGNESS TO RELOCAT'E DEBRA YOUNG: Rating __ CAREER OBJECTIVE OVERALL GPA MAJOR GPA WORK memes EXTRACURRICACTIVITY X OP EDUCA‘ITON PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE None provided 3.4 - 4.0 2.7 - 3.3 One LEVEL I and Two LEVEL 11 None 34% - 66% Above Average English plus fluent Very limited Specific 2.0 - 2.6 2.7 - 3.3 One LEVEL I and One LEVEL II Member/participant 67% - 100% Limited English plus conversant Moderately limited Broad 2.7 - 3.3 2.7 - 3.3 One LEVEL I and Three LEVEL II Officer/leader 0% - 33% Limited English only Unlimited 94 T I unmitable, not worth an interview IENNIFER CIBBONS: Rating __ suitable, might consider an interview CAREER OBJECTIVE OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVITY 96 OF EDUCATION PAID COMPUTER SELLS suitable, definite interview None provided 2.7 - 3.3 2.0 - 2.6 One LEVEL I and Three LEVEL II None 67% - 100% AM LANGUAGE SELLS WILLINGNESS TO RELOCATE JAVIO MARTIN: Rating CAREER OBJECTIVE OVERALL GPA MAJOR ORA WORK EXPERIENCE EXTRACURRIC. ACTIVITY % OF EDUCATION PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE MARY GIANT: Rating CAREER OBJECTIVE OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVITY % OI" EDUCATION PAID COMPUTER SELLS English plus conversant Moderately limited Broad 2.0 - 2.6 3.4 - 4.0 One LEVEL I andTwoLEVEL II Oflicer/leader 34% ~ 66% Above Average English only Moderately limited Specific 2.7 - 3.3 2.7 - 3.3 One LEVEL I and One LEVEL 11 None 34% - 66% LANGUAGE SELLS WILLINGNESS TO RELOCATE Above average English plus fluent Unlimited 95 F l unsuitable, not worth an interview PAUL GIBSON: Rating CAREER OBJECTIVE suitable, miyrt consider an interview OVERALL GPA MAJOR GPA WORK EXPERIENCE EX'TRACURRIC. ACTIVITY % OF EDUCATION PAID - COMPUTER SELLS exceptionally suitable, definite interview Broad 3.4 - 4.0 2.0 - 2.6 OneLEVELIandOneLEVELII Oificervleader 67% - 100% Average LANGUAGE SELLS WILLINGNESS TO RELOCAT'E BETH LANE: Rating CAREER OBJECITVE OVERALL GPA MAJOR GPA wOth DIPERIENCE EX'TRACURRIC. ACTIVITY % OF EDUCATION PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE RITA MILLER Rating CAREER oeJracrrvs OVERALL GPA MAJOR GPA wORrr EXPERIENCE EXTRACURRIC. ACTIVITY % OF EDUCATION PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE English only Very limited None provided 2.0 - 2.6 2.7 - 3.3 One LEVEL I and Three LEVEL 11 Mmbef/Pfl'fldilam 0% - 33% AW English plus fluent Very limited Specific 3.4 - 4.0 ' 3.4 - 4.0 One LEVEL I andTwoLEVEL II Member/participant 0 - 33% Limited English plus conversant Very limited r I 1 2 unsuitable, not worth an interview STAGYBROWN: Hating _ CAREER OBJECTIVE OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVTI'Y % OF EDUCATION PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE ALANOOLLINS: Rating _ CAREER OBJECTIVE OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVTTY % OF EDUCATION PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE PWOIA NOBLE: Rating __ CAREER OBJECINE OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVTI'Y % OF EDUCATION PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE exceptionally suitable, definite interview Broad 2.7 - 3.3 3.4 - 4.0 One LEVEL i and One LEVEL 11 Member/participant 0% - 33% Above Average . English plus conversant Unlimited None provided 3.4 - ’LO 2.7 - 3.5 One LEVEL 1 and Twa LEVEL [1 Officer/leader 67% - 100% Limited English plus conversant Moderately limited Broad 2.0 - 2.6 2.7 - 3.3 OneLEVEL i andOne LEVEL ii Officer/leader 0% - 33% Above Average English plus conversant Moderately limited 97 i 2 3 4 5 unsuitable, exceptionally not worth an suitable, suitable, interview might consider definite an interview interview mm unuse- Rating _ CAREER OBJECTIVE Specific OVERALL GPA 2.7 . 3.3 MAJOR GPA 2.0 - 2.6 WORK EXPERIENCE One LEVEL 1 and Three LEVEL Ii EXTRACURRIC. ACTIVTI'Y Member lpartlcipant 96 OF EDUCATION PAID 67% - 100% COMPUTER SKILLS Above average LANGUAGE SKILLS English only WILLINGNESS TO RELOCATE Very limited KEN SIMPSON: Rating __ CAREER OBJECTIVE Broad OVERALL GPA 3.4 - 4.0 MAJOR GPA 2.0 - 2.6 WORK EXPERIENCE One LEVEL 1 and One LEVEL II EX'TRACURRIC. ACTIVITY None 96 OF EDUCATION PAID 34% - 66% COMPUTER SKILLS Average LANGUAGE SKILLS English plus fluent WILLINGNESS TO RELOCATE Moderately limited ALLISON MOORE: Rating __ CAREER OBJECTIVE Specific OVERALL GPA 2.0 - 2.6 MAJOR GPA 2.0 - 2.6 WORKEXPERIENCE OneLEVELI and'NoLEVELII EXTRACURRIC. ACTIVITY Member/participant 96 OF EDUCATION PAID 34% - 66% COMPUTER SKILLS Above average LANGUAGE SKILLS English only WILLINGNESS TO RELOCATE Unlimited unsuiltabie, modesrateiy 4 exceptionally not worth an suitable, suitable, iruerview might consider definite an interview rntervrew BRIAN LEONARD: Rating __ CAREER OBJECTIVE Broad OVERALL GPA 2.0 - 2.6 MAJOR GPA 3.4 - 4.0 WORK EXPERIENCE One LEVEL I and No LEVEL II EXTRACURRIC ACI'iVTI'Y None 96 OF EDUCATION PAID 6796 40096 COMPUTER SELLS Average LANGUAGE SKILLS English plus conversant WILLINGNESS TO RELOCATE Very limited TOM WILIJAMS: Rating __ CAREER OBJECTIVE None provided OVERALL GPA 2.7 - 3.3 MAJOR GPA 3.4 - 4.0 WORK EXPERIENCE One LEVEL I and One LEVEL i1 EXTRACURRIC. ACTIVTIY Member/participant 96 OF EDUCATION PAID 3496 -6696 COMPUTER SELLS Above Average LANGUAGE SKILLS English plus conversant WILLINGNESS TO RELOCA‘I'E Very limited Ell UNDERIIIJ; Rating _ CAREER OBJECTIVE None provided OVERALL GPA 2.0 - 26 MAJOR GPA 2.0 - 2.6 WORKEKPERIENCE OneLEVELIandOneLEVELII EITRACURRIC. ACTIVITY None 96 OF EDUCATION PAID 0% - 3396 COMPUTER SELLS Limited LANGUAGE SELLS English only WILLINGNESS TO RELOCATE Very limited 99 T 1 unsuitable, not wonh an interview CRAIGTIIOMAS: Rating _ 1 3 moderately suitable, might consider an interview CAREER OBJECTIVE OVERALL GPA MAJOR GPA - WORK EXPERIENCE EXTRACURRIC. ACTIVTTY 96 OF EDUCATTON PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE STEVE HALL: Rating _ CAREER OBJECTIVE OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACITVITY % OF EDUCATION PAID comm mus exceptionally suitable, definite interview Broad 2.7 - 3.3 2.7 - 3.3 One LEVEL I and Three LEVEL ii None 34%- 66% Limited English plus fluent Moderately limited None provided 2.7 - 3.3 3.4 - 4.0 One LEVEL I and Two LEVEL ll Officer/leader 3496 - 6696 Average LANGUAGE SELLS WILLINGNESS TO RELOCATE EDITH HUNTER: Rating CAREER OBJECTIVE OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVTTY % OF EDUCATTON PAID COMPUTER SELLS English plus conversant Unlimited None provided 2.0 - 2.6 3.4 - 4.0 One LEVEL Iand‘l’hreeLEVEL iI None 096 - 3396 LANGUAGE SELLS WILLINGNESS TO RELOCATE Average English only Unlimited 100 If I l 2 unsuitable, not worth an interview CHARLES KELLY: Rating __ CAREER OBJECTIVE f 3 suitable, might consider an interview OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVITY 96 OF EDUCATTON PAID COMPUTER SELLS LANGUAGE SELS WILLINGNESS TO RELOCATE MARIA ELIJOT: Rating CAREER OBJECITVE OVERALL GPA MAJOR GPA WORK EXPERIENCE I I 4 S exceptionally suitable, definite interview Specific 3.4 - 4.0 2.7 . 3.3 One LEVEL I and‘hvo LEVEL ii None 096 - 3396 Above Average English plus conversant Moderately limited Broad 2.7 - 3.3 2.7 - 3.3 One LEVEL 1 and Two LEVEL Ii EXTRACURRIC. ACTIVITY 96 OT EDUCATTON PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE BILL OLSEN: Rating CAREER OBJECTIVE OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVITY % OP EDUCATTON PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE Member/participant 6796 - 10096 Limited English plus fluent Very limited None provided 3.4 - 4.0 2.7 - 3.3 One LEVEL I and One LEVEL II Member/participant 3496 - 6696 Am English only . Unlimited 101 I I 1 2 unsuitable, not worth an interview KIMBERLY HILL: Rating CAREER OBJECTIVE 3 suitable, might consider an interview OVERALL GPA MAJOR CPA WORK EXPERIENCE EXTRACURRIC. ACITVITY % OF EDUCATTON PAID COMPUTER SELLS exceptionally suitable, definite interview Specific 2.0 - 2.6 2.7 - 3.3 One LEVEL I andThreeLEVEL II Ofl‘rcer/leader 3496 - 6696 Average LANGUAGE SELLS WILLINGNESS TO RELOCATE ALICE ROBBINS: Rating __ CAREER OBJECTIVE OVERALL CPA MAJOR CPA WORK EXPERIENCE EX'TRACURRIC. ACTIVITY 96 OF EDUCATION PAID COMPUTER SELLS English plus conversant Very limited Broad 2.0 - 2.6 2.0 - 2.6 One LEVEL 1 and No LEVEL 11 Member/participant 6796 - 10096 Average LANGUAGE SELLS WILLINGNESS TO RELOCATE NICOLEFERRIS: Rating CAREER OBJECTIVE OVERALL GPA MAJOR CPA WORK EXPERIENCE EXTRACURRIC. ACTIVITY % OF EDUCATION PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE English plus fluent Unlimited None provided 2.7 - 3.3 2.0 - 2.6 One LEVEL 1 and Two LEVEL II Officer/leader 096 - 3396 Limited English plus fluent Moderately limited 102 r 1 unwitabie, not worth an interview MIKE iliCiillS: Rating CAREER OBJECTIVE I 3 moderately suitable, might consider an interview OVERALL GPA MAJOR CPA WORK EXPERIENCE EXTRACURRIC. ACITVTTY % OF EDUCATTON PAID COMPUTER SELLS exceptionally suitable, definite interview Broad 3.4 - 4.0 3.4 - 4.0 One LEVEL I and Three LEVEL 11 Member/participant 096 - 3396 Average LANGUAGE SELLS WILLINGNESS TO RELOCATE CINDY IIEN: Rating _ CAREER OBJECTIVE OVERALL CPA MAJOR CPA WORK EXPERIENCE EX'IRACURRIC. ACTIVITY % OT EDUCATTON PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE RICK JOHNSON: Rating __ CAREER OBJECTIVE OVERALL GPA MAJOR CPA WORK EXPERIENCE EXTRACURRICACTIVTTY % OF EDUCATION PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE English only Moderately limited Specific 2.7 - 3.3 3.4 - 4.0 One LEVEL 1 and One LEVEL ii None 6796 - 10096 Limited English only Very limited Broad 2.7 - 3.3 2.7 - 3.3 One LEVEL 1 and Two LEVEL 11 None 6796 - 10096 Above Average English plus conversant Unlimited 103 I T l 2 unsuitable, not worth an interview SUSANATENS: Rating __ CAREER OBJECTIVE T 3 moderately suitable. might consider an interview OVERALL GPA MAJOR GPA WORK EXPERIENCE EXTRACURRIC. ACTIVITY % OF EDUCATION PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE ANITA DANIELS: Rating __ CAREER OBJECITVE OVERALL GPA MAJOR CPA WORK EXPERIENCE EXTRACURRICACITVIIY % OP EDUCATTON PAID COMPUTER SELLS LANGUAGE SELLS WILLINGNESS TO RELOCATE FMNKOAVIS: Rating _ CAREER OBJECTIVE OVERALL GPA. MAJOR CPA WORK EXPERIENCE ‘ EX'IRACURRICACITVTTY % OF EDUCATION PAID COMPUTER SELLS None provided 2.0 - 2.6 2.0 - 2.6 One LEVEL I and One LEVEL [1 Oflicer/Ieader 3496 - 6696 Limited English plus conversant - --Very limited Specific 3.4 - 4,0 3.4 - 4.0 One LEVEL 1 andThreeLEVEL II Officer/leader 3496 - 6696 Limited English plus fluent Moderately limited Broad 2.0 - 2.6 3.4 - «to One LEVEL 1 and 'No LEVEL 11 None 3496 - 6696 Aboveaverage LANGUAGE SELLS WILLINGNESS TO RELOCATE English plus fluent Very limited 104 [—fi t T T I Wide 2 marina, ‘ miter, nm worth an suitable, suitable, ' interview might consider definite an interview interview IOiiN BENSON: Rating CAREER OBJECTIVE None provided OVERALL GPA 2.7 - 3.3 MAJOR GPA 2.7 - 3.3 WORK EXPERIENCE One LEVEL 1 and Three LEVEL 11 EKTRACURRIC. ACTIVTI'Y Member/participant 96 OF EDUCATION PAID 6796 - 10096 COMPUTER SELLS Above average LANGUAGE SKILLS English plus conversant WILLINGNESS TO RELOCATE Moderately limited ROBERT FISHER: Rating __ CAREER OBJECTIVE Specific OVERALL GPA 2.7 - 3.3 MAJOR GPA 2.0 - 2.6 WORK EXPERIENCE One LEVEL I and Two LEVEL 11 EXTRACIIRRIC. ACTIVTI'Y Member/participant 96 OF EDUCATION PAID 096 - 33% COMPUTER SELLS Average LANGUAGE SKILLS English plus conversant WILLINGNESS TO RELOCATE Unlimited PATIOCK WEST: Rating __ CAREER OBJECTIVE Broad OVERALL GPA 3.4 - 4.0 MAJOR GPA 2.7 - 3.3 WORKEKPERIENCE OneLEVEL I and OneLEVEL II EXTRACURRIC. ACTIVITY Officer/leader 96 OF EDUCATION PAID 3496 - 6696 COMPUTERSKILLS Limited LANGUAGE SELLS English only WILLINGNESS TO RELOCATE Moderately limited 105 'ART 3 - INFORMATION USE QUESTIONNAIRE te following questions askyou about the ways you used the information born the applicant profiles and about the ratings decisions you made regarding the applicants Please do not go back to look over your ratings while answering these restions. We are simply interested in how you believe you used the information and made your decisions. There are NO ght or wrong answers to these questions Rather, the questions will help us to understand the strategies that recruiters a: when evaluating potential J‘ob interviewees. 1. Please answer each of the following questions using the rating scale below. Place the appropriate rating to the left of each question. I 2 3 4 S to a to a to a to a to a very minimal small moderate large very great extent extent extent extent extent A. To what extent did you use the information from the applicant profile sheets in a W manner to make your ratings? in other words, to what extent did you use the same factors for all applicants (versus different factors for different applicants)? B. To what extent did you distinguish the applicants with your ratings (versus rating the applicants as pretty much the same)? C. To what extent do you believe that you are ftrlly aware of (and could accurately report) the applicant factors that were important to your ratings? 2. Consider how you used the information from the profiles to make the applicant suitability ratings. Please distribute 100 points among the faCtors listed below so that the distribution reflects the importance of each factor for your suitability ratings For example, if you believe that all of the factors were equally itnportant to your ratings, you would disuibute the 100 points as evenly as possible across the factors. Ifyou believe thatyou ignored some factors and used the others to different degrees, assign the points showing 0 points for some and varying points for the others 3 appropriate _ Remember, when you are fin'shed your points add up to 100. APPLICANT FACTORS Points (1) Oman GPA (2) Major GPA (3) Relevant Work Experience (4) Gender (5) Percentage of Education Paid (6) Willingness to Relocate (‘7) Extracurricular Activities (8) Computer Skills (9) Wm (10) CareerObJ'ective (Total 100 points) 106 3. Whenyouwereiookingovertheappilcantprofiles,doyoubelievethattherewasanythingunusualorodd aboutanyoftheapplicants? (pleasedonotlookbadtattheprofiies) _NO YES ifyou answered yes, please explain what you thought was unusual. THANK YOU VERY MUCH FOR YOUR HELP! ’iease place all materials in the attached return envelope, seal it and return it to the Kellogg Center front deskgc to the 'ront desk at Career Development and Placement Services gr: mail it by November 30, 1992 (the return envelope is rroperlyaddressedand includes all necessarypostage). Ifyou are interested in the results of this study and would like to become eligible for the prize drawings, pleaseprovideyour name andaddressbelow,or includeyour businesscard Name: Street: City: State: Zip: APPENDIX B 108 INDIVIDUAL RECRUITER POLICIES ID CO OGPA MGPA WE EA PEP CS LS WTR GEN 1 -0.1 2.5 1.3 4.6 7.2 0.1 -0.2 -.01 83.8 1.0 2 0.0 26.2 9.0 0.0 3.4 -0.1 1.3 3.7 49.2 7.3 3 0.8 0.3 60.4 3.9 27.7 1.4 -0.1 1.4 1.4 2.9 4 0.6 -0.6 2.0 -0.3 22.6 0.6 -0.2 0.8 73.9 0.7 5 61.3 4.8 0.0 11.8 13.7 1.2 -0.1 2.7 0.0 4.6 6 0.5 -0.2 8.5 12.1 22.9 11.2 -.04 0.0 38.4 7.1 7 0.0 1.9 0.3 6.4 2.5 -0.1 1.6 0.0 87.3 0.0 8 -0.1 4.2 44.0 24.5 5.2 0.0 0.9 0.4 18.9 2.0 9 -0.1 3.8 9.8 1.1 2.3 0.0 1.7 0.9 74.6 3.1 10 1.5 -0.5 0.6 17.7 -0.1 -0.2 3.2 0.5 76.8 0.6 11 0.6 4.1 5.0 10.3 72.4 0.1 1.9 3.1 2.1 0.3 12 0.5 12.7 1.1 37.1 43.1 0.9 1.8 0.0 0.1 2.8 13 9.2 -0.3 3.6 14.1 7.1 0.1 0.4 2.5 63.5 0.2 14 3.3 15.9 10.3 37.7 3.5 4.2 1.7 3.1 3.8 16. 15 7.4 1.5 8.4 6.1 22.8 0.0 1.5 4.0 43.5 4.9 16 0.0 3.6 4.3 3.7 -0.2 0.8 1.0 3.7 82.8 0.2 17 1.2 0.0 2.2 23.8 71.3 1.0 -0.2 0.8 -0.1 0.1 18 20.1 7.9 8.6 5.3 9.6 3.9 4.5 1.4 38.6 0.0 19 7.8 0.7 10.7 7.8 44.9 23.5 0.5 0.8 2.4 0.9 20 6.8 1.1 20.8 2.2 57.2 8.1 0.7 0.5 2.6 0.0 21 1.8 5.3 0.8 39.6 43.7 0.1 0.0 7.3 0.0 0.1 22 2.9 2.0 3.0 5.6 8.5 1.4 0.2 0.2 76.4 0.1 23 0.0 7.4 11.3 2.6 9.1 -0.3 1.7 -0.1 64.5 3.8 24 45.6 1.3 6.1 21.6 13.0 2.9 1.3 0.8 3.1 4.4 25 23.9 33.5 0.6 15.0 5.2 3.8 0.3 8.1 9.4 0.3 26 1.5 2.3 24.5 0.6 25.2 -0.4 -0.3 -0.1 46.4 0.4 27 39.0 -0.1 0.2 9.8 13.6 -0.2 1.4 0.8 37.4 0.0 28 77.2 2.5 14.1 -0.1 0.2 0.1 0.1 2.8 1.9 1.2 109 ID CO OGPA MGPA WE EA PEP CS LS WTR GEN 29 55.9 2.8 0.0 0.4 1.0 2.9 -0.3 0.1 36.2 0.8 30 3.6 10.6 7.1 15.5 30.6 1.5 8.0 7.6 15.5 0.0 31 0.9 1.2 4.4 67.1 6.1 2.0 0.0 16.8 2.0 0.6 32 14.7 -0.7 1.0 21.8 1.7 2.0 -0.3 0.4 59.7 0.0 33 1.9 1.3 2.8 2.3 6.4 -0.4 1.2 0.2 83.1 1.2 34 53.5 6.9 4.7 0.6 0.4 0.6 0.9 18.0 12.4 2.0 35 2.0 35.7 35.0 9.3 1.9 6.3 0.0 1.0 8.3 0.6 36 0.7 12.1 0.7 1.2 0.4 0.0 -0.7 2.1 81.0 2.6 37 0.5 4.0 20.2 5.0 21.6 11.2 5.7 10.4 18.5 3.1 38 1.2 0.2 0.3 5.7 -0.2 1.3 82.5 0.4 9.0 0.0 39 -0.1 24.1 8.7 7.3 13.0 0.0 6.0 21.1 9.6 10. 40 27.3 23.4 4.8 2.4 2.0 0.6 -0.2 2.6 34.6 0.2 41 -0.4 -0.1 12.0 27.8 36.8 4.6 2.3 3.2 11.0 2.9 42 -0.3 8.1 47.5 10.8 26.5 0.7 1.9 1.1 -0.1 3.9 43 0.2 3.3 0.1 8.3 7.1 1.3 -0.4 1.3 72.3 6.5 44 7.4 4.7 18.4 4.3 3.3 -0.3 0.2 0.0 60.6 1.3 45 0.7 6.8 -O.2 47.4 4.7 19.5 3.8 0.2 10.4 6.8 46 10.3 3.9 23.0 14.0 28.9 -0.2 0.3 0.5 17.4 1.8 47 1.1 5.5 0.0 74.4 0.5 15.0 0.0 -0.3 3.5 0.3 48 2.4 3.6 5.2 81.0 0.2 0.7 0.3 0.6 2.3 3.7 49 2.1 17.8 32.8 18.5 1.4 11.7 2.4 1.1 11.0 1.2 50 2.5 0.2 0.4 35.8 54.4 3.4 2.8 0.0 1.0 0.0 51 0.0 5.5 4.7 25.2 10.0 2.4 -0.1 0.1 50.1 2.1 110 INDIVIDUAL RECRUITER MULTIPLE R2 FROM THE POLICY CAPTURING ANALYSIS fl RATER R? RATER R2 H 1 .839 26 .684 n 2 .680 27 .725 I 3 .752 28 .628 4 .791 29 .748 5 .718 30 .720 I 6 .701 31 .762 7 .685 32 .699 8 .683 33 .855 9 .718 34 .750 10 .768 35 .797 11 .719 36 .842 12 .724 37 .474 13 .528 38 .593 14 .620 39 .502 15 .794 40 .834 16 .893 41 .778 17 .754 42 .693 18 .770 43 .750 19 .604 44 .794 20 .727 45 .581 21 .756 46 .741 22 .644 47 .802 23 .780 48 .745 24 .654 49 .723 25 .493 50 .885 51 .695 111 SPEARMAN RANK-ORDER CORRELATIONS BETWEEN THE STATISTICAL AND SUBJECTIVE WEIGHTS RATER Rho signif . RATER Rho signif . 1 .716 .010 26 .786 .004 2 .219 .272 27 .801 .003 3 .421 .113 28 .388 .134 4 .254 .239 29 .150 .339 5 .590 .036 30 .749 .006 6 .707 .011 31 .594 .035 7 .342 .167 32 .825 .002 8 .782 .004 33 .578 .040 9 .241 .252 34 .235 .257 10 .312 .190 35 .739 .007 11 .855 .001 36 .450 .096 12 .398 .128 37 .130 .360 13 .643 .023 38 .382 .138 14 .556 .047 39 .352 .159 15 .513 .065 40 .367 .149 16 .794 .003 41 .697 .012 17 .669 .017 42 .669 .017 18 .516 .063 43 .650 .021 19 .663 .018 44 .765 .005 20 .700 .012 45 .651 .021 21 .809 .002 46 .723 .009 22 .739 .007 47 .703 .012 23 .225 .266 48 .404 .124 24 .749 .006 49 .715 .010 25 -.031 .466 50 .675 .016 MICHIGAN RT V. LIB' \mmm11111111111MM 312930091411 '— __——— _——