a ’ '.’. _ . € ' o .a ”-w _ .11: .113? ""1? #7 .__..AA , Lax... _<._.,,. .” .n... .— . 3.23““ -|~7,:£. a . x .w..,...~ ‘9“ ;- regs; .3- —o ‘ ‘0 -am" v . . u ‘ , . It"! ’ {-3 ,..o o 4-‘(3‘ ' .2." ‘ 9' ~90 I “a: ‘gnv- r1. , '0.- .u- < c v o . ”a c . .5.— o a o d ‘ . ‘1 ‘ .3: : v ' I: ' f I z . :3 IE" f a 1 ‘3" I\ u'ma- I P .' ‘3: 5:3}! 1 . 3-.- . l A ~ ' 13*. 4 mid 1 133:}- ;fififi. ' 3 -' '3'?" ‘ - '- :-. 3 ; .1 mm . 2 .,.. ~ «4 "m ‘rli‘it"‘ 155:5 . - ‘ v} 5*];- ~ "t ~ Jaw-#1.. . J ‘ Pf: - "w ‘ H -. r‘v £1" 3 '1 :35: ~ r v : .» .a;&*“-=E-xt ,. : A‘s'zlo..t§,-C;"}1'.rz‘l.' ‘ ' U ..:.:~:-'~EJ :Izhnhpélu’uv :utt- 1‘3“ :. .-:' '39.” fi‘ih'VF' :uma'u ‘ E ‘ 5 H ’c 1'“~r“::*~:¥:z‘éw‘lé ' 33%" ‘rl £33515"- nfi‘ 1, sh, *3‘§B?ifix‘-.izz=gi‘zi§éév§g " ' ‘ ' ” 7 ‘ "I ‘.‘f‘w:?.='i"-§3‘3°‘11" ‘ 3“? ‘ 31:]? " .-.-* 5.3. ‘ '13 'fi.‘ ‘5‘ ‘93" ‘ : ‘. i. \ ‘i‘ 1 I u :1 2"”. .~ lg'ki‘gi‘: “#1 h ‘3.“ , ‘ (“11:13: «1’31: I $3? V. 2' 9.”;‘353! “z! ‘2 y“? 5'): ‘N v ' ~ ",0“ ‘ Y 1 I ; 1“. o . 'v‘ngl“¥ ‘ R ..‘ ) 2‘.\ ‘ 1 ‘ - fi " 1.':-}3:23.\x;.1;\‘1}.u {fiifiaf ‘ u {7‘ la“?! ‘1 '2. ‘5 'i‘izzzfii'fl-.E:,"';'::E:c7§;cz~z‘ "b a: ‘ . - 1" ‘ .ggnfifllh‘j ‘4» . 'f.- - ‘tf‘ ‘ 3 a“ ‘ ' 1b”.- - 1 i’f‘r z' 135%. 1 5 o" ‘5 ' ‘ V v ' ‘ ‘ . 3.1 :‘1: My 5 - ““ -. ML- 5‘232' a ' 1"" 1n"‘§.:.ivs “11Jt§‘-A’.‘»"-L=t\1‘tii'f~" ~ . . . . .-.l fvp . t,—.~ wk‘ .\. the $1.. dumb -:;" :"1 $513213?"'hn¥';2'§?l’9§'3“51i:“=*5§%§t:€= ‘ {E N" ' 3 . . ., . o‘ 533%“: . 93-9 . ' - ‘EL-z 3111.“. .4 ' h ‘l 1:1. 13:21:: ‘ ! lawn ~~ \ m . ‘ i ‘ o q ' P 7’ :1: I . § 1 ' H . " IE. l‘wfi“. 2'. 1.1: ' ' ‘1 LT”. '2‘ 1“ I . . ~ -'E: . - ‘ 3 3gb . . "' N! .:fa‘rt'ty“"'-‘.§‘F’§LE'.'5.5?5": xY:1.3‘.“5:‘1!{s':,\u‘ ~ ~ :fi _ '2 ‘ ‘J:';‘.}.;b~$‘:,‘v'{: ‘25. ““5“ H5 [S “ Qua” 9'.“ ==11=il?:1?1i1‘;‘r-‘r"1‘ ~h;131?.a§zivazl‘téalé’zflb - ié‘Eia'i‘iW. v %1*§%2§‘!i~.2:1“2:‘ 1» géz‘éifii‘“§ai&1§3¢ 11 :yv'inngfih m ‘!='*-'~14',. ‘1». git-w“ ‘3';;':‘1v:,1".iit...xgfii‘we1a‘fi.Blfigi‘x’i :fiWW-finu . :95? *wzzamzrzw, ' ‘ ' ‘:"‘"‘ "'"“ - ‘-' I‘ "3". x' 2-: .~.-.=-: .-:‘ ' u: -- ‘ ,...~.'3 ‘-' 34“! t'. .-. 3 ‘. ~=\§‘,mufi=‘tz‘§1“‘ 23-51%?33.15311!11x::a§11-ez.;}:';z‘w*“3.9% ',~:«.:*:,’1:§¢1.é ‘19 {3 .,1 1.3; wi‘duwt‘n‘.“ lif‘i‘."""i~§§5"i\‘£'§il “a:3:"‘3‘\~s‘-m2~5=9§ififl§§tgfi§‘ ruck 1"“ ' film. ' {-34:3{32-3- {-3 ~ f‘ \; n V l 7:1,} ”V: ‘ Hakktlis‘u’» H“ ‘1‘ I we fix [.2 u‘ y ."0A v—- o. -' ""a 4-‘.. :gfil'..u "[7" t w - 1 flat}: $1“n " V? -- .‘llv‘ v.3 - ww- ‘1‘”“51' ‘ ' ‘ I .Ao—. ' 2' '1! - ‘2‘ a 0 . 7:9 .... “‘3 .- ' . . ,"~'.:'.§£ '0‘ an ' ‘ 3'. er»? “' . ‘ "a- I r' n .47.. ‘ . ;.. ~; '0‘11‘ L t -.‘. nif-Ec’ifi ' ' ".L,~: .§ ‘ v I x“ 4‘ _, . U o. ‘9‘: ~ ,. .. ‘0 I 5‘; V . . ' 4'. ‘ "49 A ‘ WM Hpq‘flhf ' n‘ ' I n-l -.. "p" . "1-..“; 1; l I“ 3-. .- ' . ‘L: 1‘ m: .-' I 3' .‘:e'..*‘»§?%'» . .w .,‘ ' 31‘}! - ' z' .‘ l-‘.-.' 1.!1'?."‘3‘1';:‘[1'.14L"' it :1. ‘5‘. is -, I‘frg‘ .9 .‘_ @wg‘ “ ‘1 9 Q‘ "‘tl' 1:1 "I "'I 1" 3 r: 'at "1' 3 i} ‘ v ‘ u’ 'o H ' d‘ndr.‘ "1 'r 4;“: l‘ ,. I P" ‘l ,, . 01”. .y’. . VI H u. g“ 43 . «3‘ a“? rf~ 4‘ ,u ' "‘- "l‘ "J I" 1' I' ' '-1“‘:'1 ‘1" "‘7‘ .‘t' '1- I .‘ I'JI‘H "W5" H u" ‘-1J‘|“’ ll“ .‘ ' ~ '1. w ‘w . " u 3 |.-‘-' . . » «HM-1 .w {4 Iv a M1 J. - Ifzvnzmm. :-:-v '3.» 1 l. 3.. .",' .' .I" ,y . , I 'l .VV‘I'I .‘V"'q‘»..o"l'.v1m‘,l)-‘ , 1.2:. .1. 1,;‘m "' :1. J... W}.- ..'r." W ”a" h‘fl'l“ "v": ‘3‘, 'g' ,‘l'H ‘1‘1‘5: ', “ “x" ‘ "(5?5 ‘llfil’n‘m‘sfut' .';"'-' H ‘ h,‘ x a: x .. «t. “a. .wz'. M .y v. w ~ .1 '''' r». tum-3w -- mu «gaff? w i.” 7“}Ir-J d ‘ ‘Zly‘utl'h'l" :1" “ ‘ ‘ . ”Llu‘f . ‘. _ ' g H n' 1;“ I‘."w'~’g"."._4, '1'". .,.|l"_,.: ‘ ' w "A "'~:"\"‘=-H=..;Jf«.»L‘NWA:"1i‘%fi§*£.l‘l§filz*k'eaig‘éz‘;WM..... ‘ H 1“", H“ '..~\;. " W" ‘ ‘ "' 1‘."- ‘ V .w H -- .3. L'. HUI c ‘ '. 9??" ‘u‘h ‘ 1am .Lfl‘ anudr-w "W ;’."s'.i'-.w' “.1.’ai'..::;z.aa_;,t,.!~“:& b 3.13 =:\i‘m.*r;.'h'r*'.U--u .‘I .11.. THESlS \ l l l W "" is; w y 1 V‘ ’. 1: ‘t l 3199..- L5“":‘$ \‘ a 7.5"?" .. " f'. «3’1” 11; ‘1'. ‘a . “V ”I 'F j! ,- LIN an. . M .9 - v J I" V, i if fin-I'd I”- f ‘5'- “inf-F y 311-.» .10., ‘ /' This is to certify that the dissertation entitled SOURCES OF INFLUENCE IN THE FORMATION OF INITIAL JOB EXPECTATIONS presented by Paula Marie Popovich has been accepted towards fulfillment of the requirements for Ph . D . degree in Psychology ajor professor Date "I 4/2 & MSU is an Affirmative Action/Equal Opportunity Institution 0-12771 ‘IVIES‘_J BEIURNING MATERIALS: Place in book drop to uaaAmas remove this checkout from .—:—- your record. FINES will be charged if book is returned after the date stamped below. 11' 2-». r ‘wr'fiareau 9". ‘1‘"? . 1:" I I ‘ .. ' . _ 1 ~_'- I '- ‘ ‘ 412 ca. g:€:%Juit E3 to. Jun: if © 1984 PAULA MARIE POPOVICH All Rights Reserved SOURCES OF INFLUENCE IN THE FORMATION OF INITIAL JOB EXPECTATIONS BY Paula Marie Popovich A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1983 ABSTRACT SOURCES OF INFLUENCE IN THE FORMATION OF INITIAL JOB EXPECTATIONS By Paula Marie Popovich The purpose of this study was to determine the content and sources of applicants' expectations about their next jobs. The contribution of the credibility of these sources to the strength of the individual's expectations and the relationship between strength of expectations and anticipated job satisfaction were also assessed. Seventy-two upper-level undergraduate students applying for interviews through the Placement Services at Michigan State University were interviewed in order to identify their expectations about their anticipated jobs. Additionally, 108 students completed questionnaires that contained a pre—determined list of 15 categories of job expectations suggested from a review of the literature. Expectations from the interviews were sorted into the 15 pre-determined categories used in the questionnaire, and new categories were created with the remaining expectations that did not fit into any of the pre-determined categories. Analysis of the interview expectations revealed that applicants expressed primarily positive expectations about Paula Marie Popovich the job. Both methods yielded two groups of expectation sources: one group consisted of individuals or events that were directly related to a work organization (recruiter, job experience, employee of the organization), and a non- organizational group which consisted of individuals not directly related to a work organization (teachers, friends, family). A significant chi-square value indicated that the organizational sources tended to be listed more often as the first source of expectation information for job applicants. A regression analysis showed that the credibility of the source of expectation information was a small but significant contributor to the strength of the expectations in both the questionnaire and the interview samples. The strength of these expectations was greater in the interview than in the questionnaire sample. However, the correlations between the strength of the participant's expectations (and their evaluative components) and anticipated job satisfac- tion showed a low, but significant, relationship in the questionnaire sample and an equally low, and non-significant relationship in the interview sample. Job expectations as applicant desires from a pre-job search period are discussed, as is the design of more accurate measures of these expectations and their sources. ACKNOWLEDGEMENTS I would like to thank the members of my dissertation committee: Drs. Mary Zalesny (chair), Norbert Kerr, Neal Schmitt and Ben Schneider. I would also like to thank the Placement Services at Michigan State University, particularly John Shingleton and Patrick Sheetz, for allowing me to contact students through their office. Finally, I would like to gratefully acknowledge the patience and support of my friends and my family, especially my parents, to whom I dedicate this dissertation. 11 TABLE OF CONTENTS LIST OF TABLES O O O O O O O C O O O C O O 0 LIST OF FIGURES O O O O O O O O O O O O O O 0 LIST OF APPENDICES O O O O O O O O O O O O 0 Chapter I. INTRODUCTION AND LITERATURE REVIEW . The Relationship Between Beliefs and Attitudes: An Expectancy-Value ApproaCh O O O O O O O O O O O O Fishbein and Ajzen . . . . . . . Vroom's Expectancy-Valence Theory. Summary. . . . . . . . . . . . . The Formation of Beliefs: A Theoretical Approach . . . . . Source Credibility and Belief Formation. The Problem of Disconfirmed Expectations Research on Initial Job Expectations The Realistic Job Preview. . Managerial Expectations. . . Career Theory and Expectation Summary. . . . . . . . . . . The Formation of Job Expectations. Source Credibility and Job Expectations. smary O O O O O O O O O O O O O 0 Research PrOpositions. . . . . . . Proposition I. . Proposition II . Proposition IIIa Proposition IIIb The Present Study. . . . . . . . . iii Page vi viii ix N mono 11 13 15 15 15 17 20 21 24 25 27 27 28 28 29 29 Page II 0 METHOD 0 O O O O O O O O O O O O O O O O O O 3 1 Participants . . . . . . . . . . . . . . . 31' Instruments. . . . . . . . . . . . . . . . 32 Questionnaire. . . . . . . . . . . . . . 32 Pre-test of the Questionnaire. . . . . . 37 InterVieW. O O O O O O O O O O I O O O O 40 Interview Training . . . . . . . . . . . 41 Procedure 0 O O O O O O O O O O O O O O O O 42 Coding of Variables. . . . . . . . . . . . 43 Job Expectation Categories . . . . . . . 44 Data Analyses. . . . . . . . . . . . . . . 47 PrOPOSition I O O O O O O O O O O O O O O 47 Proposition II . . . . . . . . . . . . . 47 Proposition IIIa . . . . . . . . . . . . 49 Proposition IIIb . . . . . . . . . . . . 50 III 0 “SULTS O O O O O O O O I O O O O O O O O O O 52 Proposition I. . . . . . . . . . . . . . . 52 Sources of Job Expectations. . . . . . . 54 Proposition II . . . . . . . . . . . . . . 56 The Questionnaire Sample . . . . . . . . 59 The Interview Sample . . . . . . . . . . 64 smary O O O O O O O O O O O O O O O O O 72 Proposition IIIa . . . . . . . . . . . . . 73 Proposition IIIb . . . . . . . . . . . . . 74 smary. O O O O O O O O O O O O O O O O O 79 IV. DISCUSSION 0 I O O O O O O O O O O O O O O O 82 Application Expectations About the Job . . 82 Sources of Expectation Information . . . . 85 Source Credibility and Expectations. . . . 86 Expectations and Job Attitudes . . . . . . 87 iv Page Implications and Future Research . . . . . 89 APPENDICES C O O O O O O O O O O O C O O O O O O O O 92 “FERENCES O O O O O O O O O O O O O O O O O O O O O 1 o 4 Table 2. 3. 4. 10. 11. 12. 13. 14. 15. LIST OF TABLES Vroom's Expectancy - Valence Theory . . . Stages of Career Development. . . . . . . Characteristics of the Participants . . . Employees' Expectations About the Occurrence of Job Features Before Joining the Company . . . . . . . . . . . . . . . Correlation Matrix of Source Credibility I tems O I O O 0 O O O O O O O O O O O O O Expectation Categories. . . . . . . . . . Frequency of Choice of Sources. . . . . . Frequencies of Organizational and Non- organizational Sources for 15 Job Expectations for the First Source Listed. Regression Analysis Summary Table for the Questionnaire Sample. . . . . . . . . Correlation Matrix of Variables in the Regression Equation in the Questionnaire sample. 0 O O O O O O O O O O O O O O O 0 Mean Perceived Probability Occurrence of Expectation Scores for Levels of the Job Dimensions in the Questionnaire Sample. Mean Perceived Probability of Occurrence of Expectation Scores for the Interaction of Source Credibility X PeOple Job Dimension in the Questionnaire Sample . . Regression Analysis Summary Table for the Interview Sample. . . . . . . . . . . Correlation Matrix of Variables in the Regression Equation in the Interview sample. 0 O O O O O O O O O O O O O O O 0. Mean Perceived Probability of Occurrence of Expectation Scores for Levels of the Data Job Dimension in the Interview Sample. . . . . . . . . . . . . . . . . . vi Page 19 33 35 39 46 53 55 58 60 61 63 65 67 68 16. 17. 18. Page Mean Perceived Probability of Occurrence of Expectation Scores for the Interaction of Source Credibility x Things Job Dimension in the Interview Sample . . . . . . 70 Mean Perceived Probability of Occurrence of Expectation Scores for the Interaction of Source Credibility x People Job Dimension in the Interview Sample . . . . . . 71 Frequencies and Characteristics of Interview Expectations . . . . . . . . . . . . 76 vii Figure 1. LIST OF FIGURES Schematic Presentation of Conceptual Framework Relating Beliefs, Attitudes, Intentions and Behaviors with Respect to a Given Object . . . . . . . . . . viii Page LI ST OF APPENDICES Appendix Page A. Questionnaire Format . . . . . . . . . . . . 93 B O Intel-View Format 0 O O O O O O O O O O O O O 9 8 C. Labels of Worker Function Levels and Their Recoded Values . . . . . . . . . . . . 102 and Worker Function Ratings for Job Types Represented in the Main Study Sample . . . . 103 ix CHAPTER I INTRODUCTION AND LITERATURE REVIEW Whenever an individual enters a new situation, s/he brings expectations or beliefs about what that experience will be like. These expectations about the situation may be formed from the individual's past experiences and information about similar events. Expectations are the individual's way of understanding his/her environment and they provide a basis upon which later beliefs and attitudes about the situation may be formed (Fishbein & Ajzen, 1975). New employees entering a job bring with them expectations about what working on that job will be like. Researchers in Organizational Behavior believe that job applicants frequently come to the job with unrealistic (i.e., inflated) expectations (Lawler, 1973; Wanous, 1980). Research on initial job expectations has shown that if these expectations are disconfirmed by actual experiences on the job, dissatisfaction with the job and turnover may result (Porter & Steers, 1973). Despite this recognition of the importance of initial job expectations and the potential problems associated with their disconfirmation, research on these expectations has been limited. Much of the existing research on job expectations has concentrated on job search and actual job experience, focusing on the effects of initial job expectations on later job satisfaction and turnover. This research has been SE CC at ind Sit 2 characterized by the presentation of pre-determined lists of expectations to new employees who have often already been on the job. Little attention has been devoted to the content of the expectations themselves, especially the expectations of individuals who have not yet taken a job or who are in jobs in areas other than business management. There has also been no consideration of where these expectations come from, although it has been assumed that the recruiter plays a large part in creating these expectations (Bray, Campbell, & Grant, 1974; Dunnette, Arvey & Banas, 1973; Ward 8 Athos, 1972). The present investigation uses an attitude theory framework to clarify the content of the job applicant's initial job expectations, determine the pre-job sources of influence on their formation, and assess the relationship between these expectations and anticipated (future) satisfaction with the job. Initial job expectations are conceptualized as the beliefs upon which later beliefs and attitudes about the job will be based. Before the role of job expectations as beliefs for later attitudes is discussed, the theoretical relationship between beliefs and attitudes will be presented. The Relationship Between Beliefs and Attitudes: An Expectancy-Value Approach Fishbein and Ajzen. Beliefs are the information an individual has about an object (i.e., person, thing, or situation). Specifically, a belief '. . . links an object s} be be 3 to some attribute" (Fishbein & Ajzen, 1975, p. 10). "The apple is bitter" is a belief statement, as is "the task is Isimple.“ Statements such as these verbalize the links made between persons (or things or situation) and their attributes, based on information an individual has about these objects. These beliefs about an object may also acquire an evaluative Component. Using a simple example, if an individual believes that "the apple is bitter," s/he may also conclude that, because bitter tastes are unpleasant, s/he does not like this apple. This negative evaluation of one apple may then become part of the individual's attitude toward all apples. Fishbein and Ajzen explain the relationship between beliefs and attitudes as follows; “A person learns or forms beliefs about an object. These beliefs influence his (sic) attitude toward the object . . . [A]ttitude is viewed as a compound in which the elements are beliefs and the affective value of the compound (i.e., attitude) is some function of the affective value of the compound beliefs.“ (1972, p. 488) Beliefs about an object may also influence an individual's intention to behave toward that object and perhaps even influence the actual behavior itself. Figure 1 shows a schematic representation of the relationship between beliefs, attitudes, behavioral intentions and actual behavior. In this model, beliefs are the ”building blocks" X .urvtflfic 0U UUGLZUH X UUmvflDC 0U UUQQQQH X UUWDQO wwflflbv. ELrvfixr-urwrmv: I£UH3IKMCAUmUCnvLCH U 303! NMWHIHHVWR .mn .a .msmfi .cmumm a saunamflm. .uomnno cm>ww m o» uommmwm as“: muoa>mzmm can unawucmucH .mmpouwuum . .wmmflamm mcfiumamm xuosmswuh Hmapmmocoo mo cofiumusmmmnm owqummom .H musmwm xomnvmmh IIIIIIIIIII mocmsdmmH a .............................. .................. --........---HP 2 HNM x pomnno on nommmmn sum3 muom>mnmm z HNM x noonno ou uommuou nuw3 mcomucmucw‘ x nomnno cumso» masseuse x pumnno “comm mmmmamm 5 upon which the relationship between attitudes and behaviors is based. .Fishbein (1963) has translated this relationship between beliefs and attitudes into an equation. In this equation, an attitude about an-object is equal to the sum of an individual's ”N" number of beliefs about that object, along with his/her "evaluative responses” to each belief. A08; biei i=1 “Belief“ (b) in the equation is defined as the ”subjective probability" that the object (o) is related to some attribute (i). An individual often has a number of beliefs about an object but very few of these beliefs actually influence his/her attitude. The number of these beliefs which are salient is thought to be held at seven (plus or minus two), which corresponds to the limited human capacities for processing information. Although the strength of belief (or the individual's subjective probability of the occurrence of that belief) is somewhat of an indicant of belief salience, it is possible that an individual may hold a belief strongly and yet it may not be salient in a particular situation. Fishbein and Ajzen (1975) have, instead, suggested that belief salience is a function of the belief hierarchy and strength, with the most salient beliefs being those that are stated first by the individual. The first few stated beliefs are usually fairly strongly held, with successively mentioned beliefs having CI 19 pr bel rat app 6 lower probabilities and therefore, contributing less to the prediction of attitudes. Fishbein and Ajzen also believe that beliefs that are the result of information presented to the individual (as in many attitude questionnaires) may be less salient than those beliefs generated through interviews with individuals. Nevertheless, after reviewing the contribution of belief importance to the prediction of attitudes, they have concluded that even though belief importance has been defined in various ways, none of the present approaches has provided evidence for a consistent relationship between the subjective estimates of belief importance and predicted attitudes. According to Fishbein and Ajzen, the above equation is based on the same contention offered by all expectancy-value theorists in their attempts to predict behavior. As early as Tolman, expectancy-value theorists have believed that '. . . people learn 'expectations,‘ i.e., beliefs that a given response will be followed by some event" (Fishbein & Ajzen, 1975, p. 30). Vroom's Expectancy-Valence Theory. From the same origins as Fishbein and Ajzen (i.e., Edwards, 1961; Tolman, 1932); Vroom, (1964) developed an expectancy model which predicted behavior in the work setting and was based on the belief that people learned expectations. Vroom's Expectancy-Valence theory is a model of rational choice. Like other expectancy theories, Vroom's approach attempts to predict behavior, particularly choice In rf he oc ck as 7 behavior, from an individual's expectations about a situation. Expectations about the situation are referred to as “expectancies" in his model. Expectancies are defined as the individual's subjective probabilities (or beliefs) that making a certain choice (e.g., the Choice of task to perform or choice of effort level within a task) will lead to a certain outcome (Campbell & Pritchard, 1976). These outcomes have affective (evaluative) values associated with them, called valences, which determine whether the outcome has a positive or negative value for the individual. Vroom presents the theory in a series of propositions about the relationship of the theory's components to one another and to one's predicted behavior (Table 1). Vroom's first proposition deals with how an individual comes to value an outcome in terms of its expected consequences. He has used this proposition, for example, to explain occupational preferences which may result in a particular choice. Vroom's second proposition deals with how these valences and expectancies are related in the individual's choice behavior. Basing this choice on the Lewinian concept of force, Vroom postulates that the strength of these motivating forces is a product of the expectancies and valences. In simple form, this proposition has been paraphrased by Campbell and Pritchard (1976) as; Motivating Force = :2 Expectancy x Valence The individual's motivation to respond, then, is viewed as a function of his/her belief about the object (or event) 8 TABLE 1 Vroom's Expectancy - Valence Theo (From Vroom, 1975, pp. 187-188) Proposition 1. "The valence of an outcome to a person is a monotonically increasing function of the algebraic sum of the products of the valences of all other outcomes and his conceptions of its instrumentality for the attainment of these other outcomes.” Proposition 2. "The force on a person to perform an act is a monotonically increasing function of the algebraic sum of products of the valences of all outcomes and the strength of his expectancies that the act will be followed by the attainment of these outcomes.“ 9 as well as the individual's affective response to that event. Summary. The Fishbein and Ajzen and the Vroom expectancy approaches both attempt to predict events from a relationship between belief and affect. They stress the importance of beliefs, or expectancies, in their predictions. The events that these models attempt to predict are subjective, anticipated events. The future events in the Fishbein and Ajzen model are attitudes, while in the Vroom model, the anticipated events are choice behaviors. Despite the importance of beliefs in both models, and in expectancy-value theories in general, there is little empirical information on the origins and formation of these beliefs or expectations. In Organizational Behavior research as well, there has been no consideration of the acquisition of expectations. Attitude theorists have been more interested in the formation of beliefs, but even here much of what has been written about the acquisition of beliefs has been primarily theoretical. The Formation of Beliefs: A Theoretical Approach Belief formation is the establishment of the link between object and attribute. This acquisition or formation of beliefs may take place through any of three possible processes, resulting in beliefs that are descriptive, inferential, or informational (Fishbein & Ajzen, 1975). Descriptive beliefs are formed through direct experience m in; 0m not 501; 11165 10 with, or observation of, the object. Tasting an apple or' performing a task will establish the links between object and attribute in the beliefs that ''the apple is bitter" and “the task is simple.“ Beliefs that are based on an individual's own experience are often very strongly held, because one's own senses are seldom doubted. Not all beliefs are formed through direct observation of an object, because the relationship between object and attribute cannot always be observed. This is true, for example, in the formation of a belief about another individual's personality characteristics. This type of belief is often inferred from previously learned relationships, by assuming that a relationship which held true in one situation will hold in another. Inferential beliefs may also be formed through what Bruner (1957) calls a ”formal coding system," which involves a subjective system of logic used to form beliefs about objects that the individual has no experience with. Most inferential beliefs can be traced back to descriptive beliefs and Fishbein and Ajzen (1975) have noted that it is rather difficult to distinguish between descriptive and inferential beliefs. Many beliefs are based on neither direct experience nor inference, but are based on information received from outside sources. These informational beliefs, however, do not necessarily use all information provided by an outside source. Fishbein and Ajzen point out that acceptance of a message from a source depends upon a number of variables. ac “P pe; the in: rec 0th of 1 the Sou: °Per WayS 11 They cite the Yale Persuasive Communication approach (Hovland, Janis & Kelly, 1954) as a model of factors to be considered in the acceptance of the information from an outside source. The Yale Persuasive Communication approach (which will be discussed in more detail later), analyzes the elements of the communication process that affect the successful communication of a message. One of the most crucial elements in communicating information to influence an individual is the source of the communication, because it is with the source that the message originates. It also represents the first possible obstacle to the successful completion of the communication process. Source Credibility and Belief Formation The level of influence of the source, and the degree of acceptance of the message by the recipient often depends upon the believability of that source. Research on persuasive communication has shown that the credibility of the source, particularly his/her presumed expertise, is most influential in affecting a change of attitude in the recipient of a persuasive message (McGuire, 1969). Factors other than presumed expertise may influence the credibility of the source. These include the recipients perception of the power, similarity and physical attractiveness of the source (Oskamp, 1977). Communication researchers have operationalized these elements of credibility in a number of ways, but two consistently used elements have been described SC EX be SO fr! if dif Pre dis The ti0: dis< 12 as "authoritativeness." (expertise or competence) and “trustworthiness" (or character), (Berlo, Lemert & Mertz, 1966; McCroskey, 1966; 1970). The trust and expertise of the source are also important to consider in the formation of beliefs, and not only in the acquisition of informational beliefs. Bem, (1970) in outlining Rokeach's theoretical perspective on beliefs, has pointed out that; '. . . every belief can be pushed back until it is seen to rest ultimately upon a basic belief in the credibil- ity of one's own sensory experience or upon a basic belief in the credibility of some external authority" (Bem, 1970, p. 5). If the credibility of a source is doubted, even if the source is one's own senses, then the belief formed from experience with that source cannot be as strongly-held as a belief formed with the confidence of having a credible source. Attempts to predict an attitude about an object from weakly-held beliefs should also be less successful than if these beliefs were strong. It also follows that strongly-held beliefs, being difficult to change, should form strongly-held attitudes. Previously formed and strongly-held beliefs which are later disconfirmed should result in some attitudinal consequences. The literature on attitudes, and also that on job expecta— tions, show support for this phenomenon and is worthy of discussion. ta SO. Ste tot dis a £1 and 13 The Problem of Disconfirmed Expectations The negative consequences of disconfirmed expectations have long been recognized by cognitive consistency theorists, who believe that individuals attempt to maintain a balance, or consistency, among cognitive elements (i.e., beliefs, attitudes, values, intentions, etc.) as well as between cognitive elements and behaviors. Aronson and Carlsmith (1962), for example, showed that for female subjects, the desire to maintain consistency between the expectations of performance on a task and the actual performance is very strong. In their work, even an individual with negative expectations about her performance on a task would, if given the chance, go as far as change good performance to poor performance to correspond with negative expectations about her performance. Their research also revealed that a disconfirmed expectation may lead to a generalized negative affect about the situation in which the expectation was disconfirmed (Carlsmith 8 Aronson, 1961). The consistency approach to understanding the impor- tance of (disconfirmed) expectations is also advocated by some organizational researchers. For example, Porter and Steers actually defined job satisfaction as ". . . the sum total of an individual's met expectations on the job” (1973, p. 169). This view is essentially the approach of discrepancy theorists, who conceptualize job satisfaction as a function of the discrepancy between "what is" on the job and "what is expected/desired” (e.g., Locke, 1968; 1976). e) ir. th ex 14 Disconfirmed expectations on the job have been cited as a reason for both job dissatisfaction (or lower satisfaction) (Wanous, 1980) and a higher probability of early turnover 1 (Porter 8 Steers, 1973). One method proposed as a potential solution to the negative affect produced from disconfirmed expectations on the job has been the use of the realistic job preview. The realistic job preview (or RJP) is a technique designed to reduce the problems of disconfirmed expectations on the job. Proponents of the RJP believe that traditional recruitment methods help to create and maintain unrealistic (inflated) expectations of the job by presenting only positive facets of the job to prospective employees. The RJP is a recruitment alternative in which both positive and negative aspects of the job are presented to applicants in an attempt to “deflate” expectations to a more realistic level (Wanous, 1980). The new employee with a more realistic conception of the job should have fewer problems associated with having his/her expectations disconfirmed on the job than an employee who has inflated job expectations. Research on the RJP, as well as in the areas of managerial expectations and career theory, has yielded useful basic information on initial job expectations. However, because these areas of research did not have initial job expectations as their main focus, there are limits to their findings which need to be expanded upon. ir ir me 614 01“. li 0: EX 15 Research on Initial Job Expectations The Realistic Job Preview. Research on the RJP has assumed that the initial expectations of its applicants are inflated (i.e., unrealistically positive). However, information provided by this research has been limited to those expectations concerning a specific job ”preview." A problem with the RJP as a source of information on initial expectations is that the expectation or belief statements provided in the preview are actually often taken from interviews with job incumbents and not applicants. The information from this research is not about what applicants may expect or believe a job to be like, but what the job incumbent believes to be the concerns of the applicant. This information may either be what the incumbent recognizes as his/her current expectations of the job or retrospective information about the incumbents' recalled initial job. expectations. Either type of information, however, is misleading since experience on the job may have altered the incumbent's perception of initial expectations about the job. Managerial Expectations. Other research on job expectations (Bray et al., 1974; Dunnette et al., 1973; Ward & Athos, 1972) has been concerned with the expectations of only those applicants accepting managerial positions. This line of research has shown that newcomers to the organization have very positive (and often unrealistic) expectations about the job. Like the RJP research, these 16 investigations have generally used pre-determined statements of expectations/beliefs, often positively-worded, such as asking the employee about an expectation that s/he “Will have a good boss" on the job (Dunnette et al., 1973). It has been noted that the presentation of such statements often creates beliefs which the individual did not previously hold (Fishbein & Ajzen, 1975). Beliefs formed as a result of this limited contact are not strongly held and would therefore be poor predictors of attitudes about the job. Another difficulty with the research on job expecta— tions is that much of it is conducted with individuals who already have accepted positions with the organization. This is a criticism that has also been leveled at the RJP research, as there is an inconsistency in the timing of the presentation of the preview across studies (Haccoun, 1979; Popovich & Wanous, 1982). It is possible that the expectations formed before making the choice of a job may be different from expectations formed after the job choice decision. Research on cognitive dissonance (Festinger, 1957), for example, has revealed that if there is an inconsistency between a belief and a behavior, an individual may attempt to restore consistency (or consonance) by changing the belief or the behavior. If an applicant has already made the choice of a job, s/he may only admit to positive expectations/beliefs about the job in order to psychologically justify the choice. In this view, it is not 17 surprising that new employees often have high expectations about the job they have chosen (Lawler, 1973; Wanous, 1980). Although it is highly unlikely that the individual forms all of his/her expectations about the job at the time a position is accepted, research on job choice has been concerned primarily with the state of the (adult) individual at the time of the actual job choice. Applications of Expectancy-Valence theory (e.g., Vroom, 1964) to job choice follow this trend, concentrating on the individual's subjective perceptions of expectancy, valence and outcome at the time of the choice behavior only. The RJP literature is also limited by its concern with the expectations of the individual only at the time of hiring and its reliance on retrospective expectations elicited from job incumbents. There has been little interest in the developmental aspect of these expectations of the job, including their formation, if and how they change as the individual matures, and the consequences of changes. Career Theory and Expectations. The developmental perspective, however, has been an interest of career theorists. Career theorists have hypothesized that career choice behaviors change (i.e., mature) as the individual ages (e.g., Dysinger, 1950; Beilin, 1955; Ginzberg, Ginzberg, Axelrad & Herma, 1951; Super, 1953). An example of the developmental perspective of career theory is Super and Bohn's (1970) stage model of the individual's self concept as applied to the choice and development of a career SE ti in OC 18 (Table 2). In each stage, the individual copes with a different aspect of the career choice and development process and approaches these areas in different ways depending upon the stage of development of the self-concept. A crucial point in this model, as in other career stage models, is late adolescence. It is as this stage (called "Reality Testing" in this particular model) that the individual makes the attempt to focus the results of childhood and adolescent exploration into an occupational (and job) choice. The needs, motives and values that an individual brings from childhood and adolescence are recognized influences on initial choice (Schein, 1980). It is this initial choice and its consequences that will also affect later stages of the individual's development, both in the maturation of the self-concept and in future job choices. As an indicant of increasing vocational maturity, Super (1955) also has found that as individuals mature, they seek out more information about occupations. This informa- tion becomes more reliable, relevant and specific as the individual gets closer to making a decision to enter an occupation (Crites, 1969). Because occupational choice includes the choice of a specific job, this information from career theorists may also be applied to job choice and expectations. Particu- larly relevant is the concern with the late adolescent 19 Table 2 Stages of Career Development (from Super and Bohn, 1970) Exploration. Childhood and adolescent development of the self-concept. Reality Testing. The transition from school to work and early work experiences. Trial and Experimentation. Attempts to implement the self-concept by staking out a career(s). Establishment. Implementing and modifying the self- concept in the middle career years. Maintenance. Preserving and continuing to implement the self-concept. ‘ Decline. New adjustments of the self-concept following termination of one's occupational role. 20 period as a key point in determining the individual's later choices and reactions in his/her career. The expectations that the individual brings from childhood and adolescence have not yet been tested and, therefore, may be particularly vulnerable to disconfirmation by the first major job experience. Those individuals who are more vocationally “mature,“ that is those who have acquired more accurate information about the job, should have more realistic expectations. They should also have formed more expectations about the job as a result of this information gathering. S ry. A major conclusion from this brief review of the research on job expectations is that these early expectations play an important role in later job satisfac- tion and retention. However, relatively little is known about these expectations. There has been no attempt to assess the content of initial job expectations, particularly for individuals at various stages of the job search process, or for those who are searching for other than business management positions. Furthermore, what is known about these expectations may have been created in part by demand characteristics inherent in the use of pre-determined positively-worded expectation statements with individuals who have already made the decision to accept a job. In view of the theoretical importance of these expectations in the formation of attitudes, and the potential problems associated with their disconfirmation, investigation of the 21 content of these expectations and where they come from is warranted. The formation of job expectations is next discussed. The Formation of Job Expectations A11 expectations may be traced to a basic belief in the credibility of a source, whether it be one's own senses, or an outside source of information. In the case of job expectations it has been assumed that recruiters are a major source of expectations about jobs. Recruiters are often the first contact job applicants have with an organization and are a primary source of information about specific aspects of a job. Research has shown that recruiters are an influential factor in the applicant's choice of a job (Rynes, Henneman & Schwab, 1980), and that various recruiter attributes may affect the applicant's job choice (e.g., Schmitt & Coyle, 1976). It remains to be seen if the information provided by the recruiter actually does influence the expectations that a job applicant forms and brings to the job, or if there are other sources that are attended to for this information. There are sources of job information other than recruiters, such as family, friends, school, and media. Vocational theorists, for example have long recognized the roles of family, school and friends in the choice of an occupation (Crites, 1969; Roe, 1957) and recent research on television and occupations (especially that concerning sex-typing in televised occupations) has shown the media 22 to be a source of influence in the perception of occupations (e.g., Seggar and Wheeler, 1973). Reviews of how individuals find jobs (Parnes, 1954; 1970) have shown that informal sources (e.g., friends and relatives) are also a major source of information about job openings. Additionally, blue- and white-collar workers have been show to differ in the source and the amount of information they sought out about jobs (Parnes, 1970): white-collar workers have a tendency to seek out information from more formal sources than do blue-collar workers. It is also likely that the information provided by these sources is also differentially influential in the formation of expectations about the job for blue and white-collar workers because of the different natures of these job categories. In recognizing the variety of sources that could influence the formation of job expectations, it follows that different sources may influence different expectations. For example, in the above statement about the differing sources of job information for blue and white-collar workers, it may be true that a reason for this different choice of sources is that the type of information needed for a blue-collar job is different than that needed for white-collar positions. If the blue-collar worker is concerned with physical danger on the job, s/he will be more interested in information about working conditions. The white-collar applicant who is assured of a relatively safe working environment may attend 23 more to a source who can provide information on advancement opportunities. It is also possible that specific expectations about the job (e.g., exact pay levels, particular job tasks) may be formed based on information from sources who are closer to the job/organization, who possess this information. More general expectations (i.e., friendly co-workers, good working conditions) may come from other sources who are more removed from the job/organization. Another possible influence on contact with a source is the individual's position in the job search process. Super (1955) has found that as the individual matures and gets closer to making an occupational choice, s/he gathers more information about the job or organization.. An individual who is closer to making the choice of a particular job, or has already made the job choice, may seek more specific information about the job from sources who are closer to the job/organization than someone who has just started the job search. The applicant who is looking for a good source of information about the job, however, must consider the credibility of that source. Regardless of the information obtained, if the source of that information is not perceived to be believable (i.e., trustworthy, expert), then the usefulness of the information for the individual is diminished. An assessment of the credibility of the sources could help to determine which of the sources is most 24 influential regarding various job expectations. Source Credibility and Job Expectations Popovich and Wanous (1982) note the importance of the role of the source in the persuasive communications to job applicants. The authors point out that although it may be true that a job applicant receives much of his/her information about the job from the recruiter, the information may not be accepted unless s/he also believes that the recruiter is an expert and/or trustworthy source. Other sources, such as employees, or even family and friends may be more influential if they are viewed as being more credible. Porter, Lawler, and Hackman (1975) support this contention, noting that little research has been undertaken regarding the relative credibility of different sources of job information. They do cite one study (Sorenson, Rhodes & Lawler, 1973) which investigated the sources of job information for students in accounting. Sorenson et a1. (1973) attempted to identify sources of the "gap" between the values of CPA firms and those of entry-level staff accountants. They found that one of the reasons for this gap was the difference between the expectations of new accountants and what they actually experienced on the job. The authors attribute the expectations to the accounting student's reliance on classroom and instructor information about the job, which is a different perspective from.what would be presented by a practitioner in the field. From this, Porter et a1. (1975) 25 propose a hierarchy of source credibility. High levels of credibility are attributed to teachers and current employees. Lower levels of credibility are attributed to organizational advertisements and recruiters. A later study (Fisher, Ilgen & Hoyer, 1979) attempted to determine applicant perceptions of source credibility by using job search vignettes that were mailed to college seniors. The sources of job information used in these stories were the interviewer, a friend, a (job) incumbent or a professor. Participants in the study read the story and rated the source on both trust and expertise. Interviewers were found to be the least credible and job incumbents the most credible sources of information about the job. In the formation of beliefs, source credibility was mentioned earlier as a factor influencing the strength of the belief.. This should also be true in the formation of job expectations. Information from more credible sources should have greater influence in the formation of initial expectations about the job. Specifically, the information from more credible sources should lead to more strongly-held expectations (i.e., having a higher perceived probability of occurrence) than information from sources who are perceived to be less credible. Summagy Initial job expectations are assumed to be, and have been shown to be, an important factor in the individual's adjustment to the job. However, there has been surprisingly 26 little research on how expectations are formed and the role of different sources of information (including, but not limited to the recruiter) in the acquisition of these expectations. Equating expectations with beliefs is a useful means of theoretically understanding how and why expectations function as cognitive elements. This link emphasizes the importance of initial expectations in the formation of later attitudes and the lack of empirical information on their formation. Specifically, information on the content of initial job expectations and the sources of these expectations is lacking. Much of what is known about job expectations is based on the presentation of pre-determined lists of expectations to (adult) individuals who have already made the decision to accept a position. The mode of presentation may have created these expectations, in part. A more accurate assessment of the content of initial job expectations might be obtained by collecting expectations from individuals about a number of aspects of a job at various points in the job search process. Information from those who are applying/interview- ing for their first major job would be most useful, since their expectations will not only influence this first job experience but also their reactions to future jobs. Determining the sources of information upon which these expectations are based should also aid in understanding initial job expectations. The source of this expectation information, and particularly the credibility of the source, 27 may be influential in determining the strength of these expectations. Factors affecting the individual's contact with a particular source of job information, such as job type and position in the job search process (e.g., pre-job choice versus post-job choice), may also affect the strength of these initial job expectations. Clarifying the content of these expectations and determining the role of various sources in their formation should also provide the basis for a more complete understanding of initial expectations about the job. Research Propositions This study attempted to explore three major propositions regarding the content and sources of initial job expectations. The first two deal with the influence of various sources of information on a job applicant's [expectations of the job. The third proposition concerns the content of these expectations and their contribution to the formation of anticipated attitudes about the job. Proposition 1. Different sources of information will influence different expectations held about a prospective job. In particular, specific expectations about the job (e.g., exact pay levels, particular job tasks) may be influenced by sources that are closer to the job and the organization (such as employees of the organization). More general expectations (e.g., friendly co-workers, high responsibility) may be influenced both by sources that are 28 close to the job or organization and by sources that are more removed from the job or organization (such as one's family). Proposition II. Information from sources perceived to be more credible should contribute to the formation of stronger expectations about the job (i.e., having a higher perceived probability of occurrence). Other factors that may affect both the source used and the strength of the expectations include: a. the individual's position in the job search process. Individuals who have committed themselves to the choice of a particular job (i.e., have accepted a job offer) should have stronger expectations in order to justify their job choice. b. the occupational area of the job. Parnes (1970) has shown that blue- and white-collar workers differ in their choice of sources of information about job openings. This difference in choice of sources of information may also be reflected in individual occupational areas, for example, engineering versus sales or personnel. Occupations that entail different work requirements (and situations) may require different information, as in the general case of blue- versus white-collar workers. Proposition IIIa. It is proposed that the expectations (or beliefs) derived from interviews with participants, being a part of the individual's actual belief system, will be more strongly-held (i.e., have higher 29 ratings of probability of occurrence on the job) than those expectations which are presented to participants from a pre-determined list. It is also proposed that the expectations elicited from participant's in interviews will be more strongly related to the individual's perceived job satisfaction than the expectations of members of the questionnaire sample. Prgposition IIIb. Expectations of job applicants that are derived from the interview method will be different in cOntent from those expectations contained in the pre- determined list of expectations. Since these pre-determined expectations are often the result of information provided by job incumbents and are generally only positively-worded, they may not be representative of the expectations of job ,applicants. The interview should elicit expectations that are more representative of the applicant's actual belief system. The Present Study The present study examined the content and sources of the initial expectations that job applicants bring to the job. In order to extend past research on job expectations, this study involved interviewing or administering questionnaires to individuals who were at various points in the job search process, from pre-job search through to post-acceptance of a job offer (without yet being on the job). Participants in the study were upper—level under- graduate students. These college students were at a point 30 in their vocational development at which they were attempting to focus the results of childhood and adolescent exploration on the choice of a job (i.e., ”reality testing"). The use of students representing various majors also provided information on a variety of jobs. Participants were assigned randomly to either a questionnaire/survey or an interview method of data collection. Participants receiving the questionnaire were presented with statements describing various aspects of a job. They were asked to rate each aspect on two evaluative scales (Good/Bad and Harmful/Beneficial). Participants were then asked to rate each job aspect concerning the probability of its occurrence on their prospective job, thus giving a measure of ”strength of expectation" for this job aspect. Sources of information on each aspect (or expectation) were identified and rated on their perceived credibility. Information on the type of job being sought, stage in the job search process, and ratings of anticipated job satisfaction were also requested. The interview format was similar to that of the questionnaire, with one exception. Participants in the interview were asked to generate their own list of job expectations as opposed to rating pre-determined expecta- tions about the job. This procedure identified additional aspects of the job about which expectations are formed but which were not included in the questionnaire. CHAPTER II METHOD Participants Participants in the study were 180 upper level undergraduate college students. The students were at various points in their search for a post graduation job, from pre- (i.e., applying/interviewing for a job) through ‘post-job acceptance (i.e., have accepted a job offer). However, none of the participants had actually started in their new positions at the time of the study. Participants included volunteers from a list of 600 students chosen randomly from all students who applied for job interviews through the Placement Services at Michigan State University from September, 1981 to May, 1982. Students were assigned randomly to either an interview or questionnaire method and were contacted through the mail with a request to partici- pate. As incentive to participate, those who volunteered were entered in a drawing for a $100 cash prize. In addition to the Placement Services sample, volun- teers were solicited from two upper-level Industrial/ Organizational psychology classes at Michigan State University. This was done to insure a variety of job types among participants. Approximately 50 of the students in these classes were considered eligible to participate (i.e., were graduating seniors who had been actively searching for jobs, and had not yet started on a new job) and were 32 were randomly assigned to either the questionnaire or inter- view method. Twenty-eight per cent (or 180) of the 650 students who were contacted participated in the study. Thirty-six percent (or 72) of the 200 students contacted for interviews agreed to participate, and 24 per cent (or 180) of the 450 questionnaire participants returned completed, usable instruments. The mean age of the sample was 22 years. Forty percent of the participants were male and 60 per cent were females. Table 3 presents characteristics of both the questionnaire and interview samples. Instruments The interview and the questionnaire developed for this study consisted of three major sections each; a. a measure of job expectations, b. identification of sources for each expectation and rating of each sources's perceived credibility, and c. job and demographic information. Questionnaire. In order to measure job applicant's expectations about their jobs, a list of categories adapted from Dunnette, Arvey and Banas (1973) was used in the questionnaire. This list of 15 job statements is brief and more representative of the major job dimensions of interest to researchers in Industrial/Organizational psychology than other established lists of expectations (e.g., Bray et al., 1974; Ward & Athos, 1972). Also it is similar to other lists of expectations in that the job statements are 33 Table 3 Characteristics of the Participants Questionnaire Interview Total Characteristic Sample Sample Sample 1. Number of Participants. n=108 n=72 n=180 2. Mean Age 22.07 21.90 22.00 3. Sex Male 42% 38% 40% Female 58% 63% 60% 4. Job Search Position Pre-job acceptance 63% 70% 66% Post-job acceptance 32% 25% 29% Other/missing 5% 5% 5% 5. Job Type a. Accounting/Finance 12% 11% 12% b. Engineering 16% 14% 15% c. Management 16% 18% 17% d. Sales 13% 17% 17% e. Advertising/Mkting 6% 7% 6% f. Social Services 5% 10% 7% g. Other/Missing 33% 24% 27% 34 positively-worded and developed primarily for use with business school graduates (Table 4). Each page of the questionnaire listed a different one of the 15 job expectations. Each expectation was followed by two sets of evaluative bi-polar adjectives (Good/Bad, Harmful/Beneficial) separated by a seven-point rating scale. These adjective pairs were taken from.Fishbein's 'AB' scale of evaluative adjectives (Fishbein & Raven, 1969). For each expectation, participants were asked to estimate the probability of occurrence of the expectation on their next job. The participants were then asked to list up to four sources of information about expectation. An option for ”No source” was included. Examples of 17 information sources were provided, although participants were permitted to list any source they might prefer. These examples were derived from interviews with individuals who were searching for jobs and from the literature (Sorenson, Rhode, 8 Lawler, 1973) (Appendix A). The questionnaire provided for ratings of perceived credibility and influence for the first two sources listed for each of the 15 job aspects. Estimates of the amount of information provided by each source were also requested. The source credibility scale consisted of five pairs of bi—polar adjectives on a seven-point rating scale (adapted from McCroskey, 1966; 1970). It is one of the few such scales that has been used in more than one experimental bUN \DQQO‘U’I 10 11 12 13 14 15 35 Table 4 Employees' Expectations About the Occurrence of Job Features Before Joining the Company (Dunnette, Arvey & Banas, 1973) Fair company policies Good salary Good working conditions High level of responsibility Recognition for good work Opportunity to use own abilities Opportunity to use own ideas Security on the job To be in charge of other peeple To have feelings of accomplishment To have high status To have interesting work To have Opportunity to get ahead To have variety on the job Will have a good boss 36 situation. A sixth pair of bi-polar adjectives (Very Influential to Me/Not at all Influential to Me) was added to assess the level of influence of the source in providing information about each expectation. The positive and negative end-points of the bi-polar credibility items were alternated to control for response sets in participant ratings. The order in which the expectations were presented in the questionnaire was randomized to preclude order effects in the participant responses. The final page of the questionnaire elicited personal information about the participants and their job search process. Information about age and sex was requested as well as job type, work history and a rating of the importance of the job for which the participant was currently applying or had accepted. A list of steps from pre-job search through to acceptance of a job offer was included to determine the participant's position in his/her job search. Anticipated satisfaction with the next job and the participant's anticipated job satisfaction if initial job expectations were later disconfirmed were measured using the Faces scale (Kunin, 1955) with the male and female faces (Dunham & Hermann, 1975). This part of the questionnaire also contained a measure of how certain the participant was in his/her expectations about the job (see Appendix A for questionnaire format.) 37 Pre-test of the_Questionnaire. The questionnaire developed for use in the study was pre-tested with a group of upper—level college students who were similar to the anticipated sample of students participating in the main study. This was done to insure that the questionnaire's written instructions were clear and easy to follow. It was also important to determine if the questionnaire itself was too lengthy or unrealistic in its request to rate the probability of occurrence of initial job expectations on a future job as well as to elicit sources of these expectations. Participants for the pre-test of the questionnaire were 36 volunteers from an undergraduate course in Industrial/Organizational psychology at the University of Michigan (Flint campus). Participation was in exchange for class credit. Although the mean age of this group of participants (28.80 years) was somewhat older than that for most college students, they were similar to the main study sample in proportion of males to females (36 percent and 64 percent respectively). Participants were given the questionnaire with instructions to complete it on their own and return it to the Experimenter within one week. Participants were also asked to comment on any problems they had in understanding or completing the questionnaire. The participant's comments on the questionnaire revealed a problem with the length of the scale used to 38 assess the credibility of the source of each expectation. The 12-item scale used in the pre-test (adapted from McCroskey, 1966; 1970) consisted of two sub-scales, which measured the ”Trust“ and "Expertise“ dimensions of source credibility. The full credibility scale was reduced in length by examining the intercorrelation matrix (Table 5) for items that were highly related and could be represented by one item. The pre-test results showed that all intercorrelations of the scale items were significant. This indicated that all 12 items were related, regardless of sub-scale. To reduce the scale length, one item was chosen to represent each of the sub-scales, although the final scale was not designed to be broken into sub-scales for the actual study. Item 10 (Awful/Nice) was chosen as most highly intercorrelated with items in the "Trust! dimension of source credibility and item 11 (Expert/Inexpert) was selected to represent the ”Expertise" sub-scale. Several of the items in the full credibility scale, such as item 7 (Intelligent/Unintelligent) and item 12 (Virtuous/Sinful) were not considered for the reduced scale. The decision to elimdnate these items was based on pre-test participant complaints that these items were not appropriate for rating the credibility of various pe0ple, situational and information sources of expectations. Items 4, 6, and 8, although correlated significantly with the other items, were not correlated as highly with either the "Trust" or 39 manuals—5m sumuuuooxmws a. .oncomunsm sum-Eu: + ..mc.V3 accozdcuam was mcofi—uumuuoo :4 c Houcam + s. so as am a. an «M en fin a. an .on.~. nn.m \usosuus> .Ns uuooxosn a. a. as am am .n .u an mm c. a. .s~.~. ~e.m \uumaxu .3” oosz + mg .m In as am am «4 an ~q .o~.~. .a.m \Hsuza .os DOOHSDH03 as am mm mm a. a. as mm cm .-.H. oa.m \oaausfia> .m acquaumca + Nm an a. on on .m N. .ms.~. os.m \auqusom .o unmoAHAoucsca .. hm em em am am am lss.s. am.m \ucooawdmucu .s accwnuansa + a~ as am pm. SN mc.H. ps.s \ucuouwsa .s cosuasoso .. ~n ~o mm a. .mv.s. sm.m casuannauca .m saccosuuca + 0. ~. mm .NH.~. aH.s \sducusum .v GOEHOunficb «s cm no .m~.H. nH.s \cmauou:~ .n amuse: + as .-.H. sa.m \auuzoamso .~ «Handguns: .. xx .-.~. ma.w \odnuqaaz ._ as as a a a s m s n w a ..>oo qcam. museum «soua mason auaadaaumuo com: «mason madam Nuddandomuu mousom mo Nahum: codunamuuou m manob hufldqnficmuu 40 "Expertise" sub-scale (or with each other) as were the others variables. Because these items were part of the established scale and could not be adequately represented by either item 10 or 11, they were included in the reduced scale as measures of other aspects of source credibility. Interview. With one exception, the interview was similar to the questionnaire in content and form. Unlike the questionnaire, which presented participants with a list of expectations, the interview method allowed participants to generate their own list of expectations about their prospective jobs during a brief telephone conversation with an interviewer. During the interview, participants were asked to think about what they expected their next job to be like and to list, verbally to the interviewer, their expectations. If the participant could not generate a list of expectations with this approach, the interviewer then asked the participant to describe the job, including possible duties. In the event that the participant was still unclear about the purpose of the interview, the interviewer provided the participant with general aspects of the job (e.g., working conditions, co-workers) as prompts. The participant was also given the opportunity to express ”no expectations” in the interview fommat. (See Appendix B for the interview format.) After the expectations were listed by the participant, the interviewer transferred the expectations to a question- naire which was identical to that used in the questionnaire 41 method. The questionnaire was then mailed to the interview participant to be completed and returned. As in the questionnaire method, the participant rated each expectation on two evaluative scales, listed sources of this expectation information, and rated the perceived credibility of the sources. Job and personal information were also obtained. Participants who had no expectations about the job completed a shorter version of the questionnaire that contained questions about the type of job being applied for, position in the job search process and anticipated job satisfaction. Interviewer Training. Three undergraduate students served as interviewers in this study in exchange for independent study credit. The training of these students took place over three sessions, each of which lasted approximately one and one-half to two hours. The first session involved a discussion of the general purpose and procedure of the study. The format of the interview was covered and modeled by the Experimenter. The interviewers were then instructed to conduct at least five interviews with friends who were searching for jobs. Session two consisted of a discussion of the results of the interviewers' practice interviews and of revision of the interview format. Interviewers were encouraged to practice the interview before the final session. Session three was a "test” session in which each interviewer conducted an interview with a single student volunteer in the presence of the Experimenter and the other interviewers. 42 Reliability estimates of the interviewer's performance were assessed by taping an actual telephone interview with each of the interviewers. The information from the tapes was then coded by each of the other interviewers. Since the interview phase of the study was rather brief (two weeks) reliability was assessed once, from interviews taped during the middle of the interview phase. The average agreement among the interviews on the expectations listed by the participants in these cases was 94 per cent. Procedure The names of 600 college seniors were selected randomly from an alphabetically-ordered list of 3,700 students who had applied for interviews through the Placement Services at Michigan State University from September, 1981 to May, 1982. The selection was made by choosing every sixth name of the list. From the list of 600 names, 400 students were chosen . randomly (by selecting the first two of every group of three names on the list) to participate in the questionnaire method. The questionnaire and a cover letter were mailed directly to the participants. The letter explained briefly the purpose of the study and informed the participant about the drawing for a cash prize for their participation. Two upper-level undergraduate classes at Michigan State were also solicited for eligible volunteers to participate. Questionnaires and cover letter were presented directly to interested students. Those students who wished to participate were instructed to return the questionnaire in 43 the envelope that was provided. Complete instructions for the questionnaire were included. The remaining 200 students from the sample were mailed a letter informing them that they had been chosen to participate in an interview concerning the expectations of their next jobs. Those students who wished to participate in the interview were asked to contact the Experimenter (of which four did) or wait until contacted (by telephone) by one of the interviewers. One-hundred eleven of those students participated in a brief (lo-minute) interview (10% of the participants who were contacted refused to participate). A follow-up questionnaire, containing the expectations elicited from the interview, was sent to each participant to determine the sources of this information as well as job and personal information. Sixty-five percent (or 72) of the participants completed and returned useable questionnaires. In both the questionnaire and the interview method, a post-card or letter reminding the participant to complete and return the questionnaire was mailed to participants one week after the questionnaire had been sent out. Coding of Variables I A codebook was developed for quantifying and scoring the variables in both the questionnaire and interview methods. Along with the Experimenter, two undergraduate students, working for independent study credit, served as coders. The coders, who had also participated in the study as interviewers, were trained in two sessions. The first 44 meeting involved discussion of the variables to be coded and practice in coding completed questionnaires. The coders were then required to complete a case between training sessions. The second session consisted of scoring the practice case and a discussion of any problems. Reliability estimates were made by having each coder score one case from each of the other coders. The average agreement of coders across the variables in these three cases was 98 per cent. Job Expectation Categories. Twenty-four volunteers from an undergraduate course in Industrial/Organizational psychology at Michigan State university sorted the job expectations that were elicited from the interviews into categories. Participation was in exchange for class credit. - The mean age of the participants was 21 years; 34 per cent of the volunteers were males. Interviews with job applicants yielded 347 expectations about the job, which were reduced in number before being presented to participant judges for sorting into categories. This reduction was accomplished by first condensing the similarly—worded items and removing those items which could be placed into the 15 pre-determined expectation categories used in the questionnaire. The resulting list of 159 expectations consisted of those statements which the Experimenter judged as not fitting into the existing 15 categories. Six of the 24 participant judges sorted as many as possible of these 159 expectations into the 15 pre- 45 determined categories. The remaining expectations were then sorted into new categories which were labeled by the judges. The resulting group of job statements were further grouped into 42 categories by the experimenter (Table 6). These categories included the original 15 pre-determined expectations from the questionnaire plus negative options for each of these expectations (the positive categories represented both neutral and positive expectations). Changes were made in three of the pre-determined categories to more clearly reflect the expectations found in the interviews. The category "Opportunity to use own abilities" was modified to include "skills" along with abilities, and ”benefits” was added to the ”Good Salary" category. ”Good working conditions“ was divided into three categories (each with a negative option). These new categories were; physical working conditions, work environment/climate, and work load/schedule. Additional categories included; Opportunity for training and experience No opportunity for training and experience Good interpersonal relationships (with co-workers 5 customers) Poor interpersonal relationships Positive adjustment/relocation experiences Negative adjustment/relocation experiences Challenge The 159 expectations from the interviews and the 42 new categories of expectations were then presented to the 01 IL 02 17 03 18 39 46 Table 6 Expectation Categories Fair company policies Unfair company policies Good salary (and benefits) Poor salary Good physical working conditions Poor physical working conditions Good work load/schedule 40 Poor work load/schedule 35 Good organizational atmosphere (positive climate) 36 Poor organizational atmosphere (negative climate) 04 High level of responsibility 19 Low level of responsibility 55‘ Opportunity to use own (skills and) abilities 20 Limited or no opportunity to use own skills and abilities 06 Opportunity to use own ideas 21 Limited or no opportunity to use own ideas 07 Recognition for good work 22 08 _2_§ 09 24 10 25 11 Limited or no recognition forggood work Security on the job Limited or no security in the job To be in charge of other pe0ple Limited or no opportunity to be in charge of other gpegple To have feelings of accomplishment To have little or no feelings of accomplishment To have high status 26 To have low status 12 To have interesting work 27 To have work that is not interesting 13 To have opportunity to get ahead 28 To have little or no opportunity to get ahead 14 To have variety on the job 22 15 30 “‘fl°””1 u: _qum b To have little or no yariety on the job Will have a good boss Will have agpoor boos Positive training/learning experiences Negative training/learning experiences Good interpersonal relations (with co-workers & customers Poor interpersonal relations (with co-workers & customers Positive relocation/adjustment experiences Negative relocation/adjustment experiences Tasks specific to a particular job Challenge 47 remaining 18 of the 24 judges. The participant judges were instructed to assign each of the 159 expectations to a category. The percentage of agreement between the judges for the categorization of each expectation was computed and the expectation was then assigned to the most agreed upon category. The average agreement over all expectations among the participant judges was 53 percent. Although this number is low and may be biased in favor of some categories which were chosen more often than others, the nominal nature of the data as well as the large number of categories precluded the use of other methods to assess the inter-judge reliability. Data Analysis Proposition 1. It was proposed that different sources of information about a job would be associated with different expectations held by job applicants. To test this proposition, frequencies of occurrence were determined for each of the expectation categories. Frequencies of occurrence for each of the sources of this expectation information were also assessed for both the questionnaire and the interview samples. In order to determine whether the choice of source was related to any particular subset of the expectation, chi-square tests of the relationship between sources of information and job expectations were performed. Proposition II. Secondly, it was proposed that the probability of occurrence of job applicants' expectations« 48 would be a function of the credibility of the source of this expectation information as well as other factors which could influence the applicant's contact with a particular source. ‘ Regression analysis was used to assess the influence of the credibility of these sources on the strength of the job expectation. The dependent variable in the equation was the strength of the participants' expectations about their next jobs. This was measured by assessing the perceived probability of its occurrence on the job. Independent variables in the equation included the credibility of the primary (i.e., first listed) source, the participant's position in his/her job search process, and the type of job being sought. The credibility of the source was measured by summing the scores for the five pairs of items in the source credibility scale, thus yielding a single credibility score. The applicant's position in their job search was measured by their response to a seven-step scale which ranged from currently not looking for a job to accepting a job offer. The job type categories (e.g., engineering, sales, etc.) consisted of three scores, representing the Data, People and Things aspects of occupations, using the dimensions from the Dictionary of Qgggpational Titles (United States Employment Services, 1977), (Appendix C). Each job type was given a three part score which represented the level at which it required dealing with data, people or things. These dimensions were used in order to permit better generalization of results 49 from the jobs included in this sample to other jobs. The‘ use of these dimensions also allowed the job type variable to be represented on a continuous rating scale for the regression analysis. In order to assess the relationship between the credibility of these sources of expectation information and the job search position and type of job sought by the participant, interaction terms were calculated and included in the regression equation. These terms were created by multiplying the main effect variables of interest to yield a multiplicative interaction term. Only those interaction terms hypothesized as relevant were included. The inter- action terms of interest were; Source Credibility x Job Search Position (C X S), Source Credibility x Data component of the job (C X D), Source Credibility X People component of the job (C x P), and Source Credibility x Things component of the job (C x T). Separate regression analyses were performed for the questionnaire and the interview samples. Proposition IIIa. Thirdly, it was proposed that expectations that are derived from interviews with job applicants would be perceived as having a higher probability of occurrence and be more strongly related to anticipated job satisfaction than those ratings of expectations presented to applicants from a pre-determined list. The difference in the average strength of expectation (or probability of occurrence) ratings between the questionnaire and interview samples was tested using the 50 Student's t statistic. The degree of relationship between the strength of these expectations (and their evaluative components) and anticipated attitudes about the job was assessed using the Fishbein and Ajzen (1975) formula; 1: A =szb.e. N "A0” represents the individual's attitude about a particular object (or person, event, etc.). In this case, A0 is the participant's anticipated job satisfaction. The term 2 biei refers to the summation of the perceived probability of occurrence values for all N beliefs about an object, multiplied by their evaluative components. For this analysis,:i biei was derived by multiplying the perceived probability of occurrence for each expectation (b) by its corresponding evaluation (e). The evaluation scores were determined by summing both of the evaluation scales (Good/Bad and Harmful/Beneficial) for each expectation. These b x e scores were then summed for each participant and averaged. This was done to reduce the large discrepancy in the average number of expectations between the questionnaire and interview samples (15 and 4 respectively). A Pearson product-moment correlation was calculated in order to assess the relationship between the expectations (IE'biei) and the anticipated job attitude (A0). Separate correlations were computed for the questionnaire and the interview samples. Proposition IIIb. It was proposed further that job 51 applicant expectations derived from the interview method would be different in content from those expectations contained in the pre-determined list used in the question- naire method. To assess this proposition, the most frequently cited expectations of the interview participants were determined and were compared to the pre-determined list of expectation categories. CHAPTER III RESULTS Proposition I It was proposed that different sources of information about a job would be associated with different expectations held by job applicants. Categorization of the job expectations revealed that very few of the expectations referred to specific informa- tion about a particular job (e.g., exact pay levels, particular job duties). Only four per cent of the expectations that were elicited through interviews could be classified as "job specific." This number was felt to be too small to make any meaningful comparison of sources between ”specific" and "general” expectations about the job, as had been proposed. However, frequencies of the type of sources listed by participants across all expectations (collected from questionnaire and interview samples) revealed that the most frequently listed sources formed two major groups. One group consisted of individuals or events that were directly related to some work organization (organizational sources), and a second group consisted of individuals not directly related to a work organization (non-organizational sources). Table 7 shows the frequency of citation of the sources in 53 Table 7 Frequency of Citation of Sources Source no. 1* - Source no. 2* Source Name Quest. Int. Quest. Ingy Parents 5% 5% 5% 8% Siblings 2 2 1 3 Other Relatives 0 1 0. 0 Friends 6 8 6 9 Teachers - 11 15 8 16 School Counselors 1 l 1 2 Placement Services 1 3 1 2 Recruiter/ Interviewer 15 24 8 15 Employee of the Organization 16 12 11 7 Organizational Advertisements 1 0 2 2 Personal Experience on a Job 16 16 6 6 Experience with Job as a Customer 1 1 1 1 Television 0 0 0 0 Radio 1 1 0 0 Newspapers/Magazines 0 1 2 2 Books/Journals 3 1 1 2 Other 6 3 3 2 Questionnaire Interview No Source 17% 8% Number of Observations n=1620 n=308 * Source no. 1 refers to the source listed first by the participant Source no. 2 is the second source listed by participant 54 each method (across all expectations), by first and second listing. The three most frequently chosen sources across both the questionnaire and the interview methods and the order of choice were; recruiter/interviewer, personal experience on a job, and employee of the organization. The three next most frequently cited sources were teachers, friends and family (parents).1 1 Sources of Job Expectations. Chi-square analyses were performed to assess the relationship between (organizational and non-organizational) sources of job information and job expectations. Table 8 is a frequency table for the first source cited by participants in both the questionnaire and interview samples. The chi-square value for the relation— ship between organizational sources of information (Group 1), non-organizational sources (Group 2), and the 15 pre-determined expectations categories was significant 2 (14) in the table reveals that, with one exception (“To Have High (X = 35.81, p .05). Proposition II Secondly, it was proposed that the probability of occurrence of job applicants' expectations would be a function of the credibility of the source of this expectation information as well as other factors which could influence the contact with a particular source. Regression analyses were used to assess the influence of source credibility, job search position and job type on the strength of the individual's expectation (which was Operationalized here as the perceived probability of its 57 occurrence on the job). A single regression equation was computed for all expectations across all participants. Although it was assumed that the expectations of each participant were independent of each other, they may have been related. This possible interdependence of expectations may have led to a biased analysis, since participants who had a large number of expectations may have been more influential in the equation than those who had few expectations. However, due to the variable number of expectations per subject in the interview sample (and the necessity of performing similar analyses on both the questionnaire and interview samples for comparison purposes), tests of the interdependence of the expectation variable and the use of multivariate techniques were not feasible. In order to test the contribution of these main effect variables and the influence of interactions of these variables, a hierarchical moderated regression was performed (Kerlinger & Pedhazur, 1973). This technique involves testing for the significance of the variance of the dependent variable which is accounted for by the interaction vectors, above and beyond the variance accounted for by a linear additive model of the main effect variables. The main effect variables are entered in the equation first and the resulting R2 value is compared to the R2 of a model containing the main effects-plus-interaction terms. The difference between the two R2 values is then tested using an Table 9 Regression Analysis Summary Table for the Questionnaire Sample Dependent 2 Variable Step Variable Entered F to Enter R Overall F Probability 1 Source Credibility (C) 66.82 .055 of Occurrence of the Data (D) N.S.* N.S. Expectation Job Search Position (S) N.S. N.S. People (P) 8.65 .068 Things (T) 9.29 .076 19.09 (df = 5/1168, (p( .05) 2 C X P 21.52 .096 C X T N.S. N.S. C X S N.S. N.S. C X D N.S. N.S. 13.87 (df = 9/1164, p< .05) *N.S. = Not Significant 59 F-test. Thg_Questionnaire Sample. Table 9 presents the results of the regression analysis for the questionnaire sample. For the main effect variables, R2 = .076. For the main effects-plus interaction terms R2 was equal to .096. The difference between the R2 values for the two models was significant F = 6.79, d.f. = '4/1164, p<.05). This indicates that it is not an additive model of the main effects variables, but a main effects-plus-interaction terms model which best describes these data. In the questionnaire sample, the main effects of source credibility and the PeOple and Things dimensions contributed significantly to the strength of the participant's expectation about the job. The interaction between Source Credibility and the People job dimension (C X P) also contributed to the variance in the dependent variable, beyond that accounted for by the main effect variables. Of the main effect variables, the perceived credibility of the (first choice) source of expectation information accounted for the most variability in the probability of occurrence of the expectation (or strength of expectation) This relationship between source credibility and expectation strength is also reflected in the significant simple correlation between the variables (r = .23) (Table 10). Along with Source Credibility, the main effect variables of the People and the Things job dimensions also -contributed significantly to the perceived probability of 60 Table 10 Correlation Matrix of Variables in The Regression Equation in theQuestionnaire Sample >4 +:m >. s :18 .t’. ‘5 .33 a): (U (D g :5 msa 01: .4 '3. c 3?: m ‘3 m 00 'J'U C3 a) .C gm 0 U O H0 00 n. Ed 00 9‘0 "38 an. :“3 Variable Source Credibility 23* Data 05 —04 People 01 -22* -13 Things -13 08 35* -14 Job Search Position 11 15 17 -15* 28* Source Cred X*** - Search (CXS) 21* 58* 13 -19 27* 88 Source Cred X Data (CXD) 22* 75* 62* -25 30* 23* 53* Source Cred X People (CXP) 10 38* -16 76* -09 -04* 15 19 Source Cred. X Things (CXT) 17 25* 34* -14 97* 29* 36* 43* -02 . significant at p<'.05 ** Dependent variable 4*: Source Cred - Source credibility Mean Perceived Probability of Occurrence Table 11 of Expectation Scores for Levels of the Job Dimensions in the Questionnaire Sample Levels of Standard Number of Job Type Dimensions Mean Deviation Observations PEOPLE* 1 Taking Instructions 72.38 19.69 56 3 Speaking-Signalling 71.38 26.24 1093 4 Persuading 70.84 26.84 265 9 Mentoring 70.40 27.80 75 THINGS* 1 Handling 69.52 26.72 1228 7 Precision Work 79.29 21.85 261 *Only the levels shown were represented in the sample. 62 occurrence of the job expectation. Table 11 shows the mean perceived probability of occurrence of the expectation scores for the two levels of the Things job dimension which were represented in the sample. The means reveal that participants who were searching for jobs that involved the lower level (i.e., ”handling“) of the Things dimension appear to have less-strongly held expectations than participants whose prospective jobs involved a more advanced level of dealing with things (i.e., “precision work"). In this sample, the high level of the Things job dimension was composed almost entirely of prospective engineers, while the lower level of the dimension consisted primarily of business students searching for jobs in sales, management, adver- tising, marketing, etc. It is those participants who were looking for engineering positions who appeared to have the strongest expectations about their future jobs. The contribution of the People job dimension may be seen most clearly in its relationship to the credibility of the source of expectation information, specifically in the significant interaction between Source Credibility and the Pe0ple job dimension (C X P). Table 12 shows the mean perceived probability of occurrence of expectations ratings for the levels of the Pe0ple job dimension which were presented to this sample, and the ratings of source credibility by the participants at each level of this job dimension. In order to facilitate the interpretability of the table, the source credibility ratings, which have a 63 Table 12 Mean Perceived Probability of Occurrence of Expectation Scores for the Interaction of Source Credibility X People Job Dimension in the Questionnaire Sample Source Credibility low high People Job Dimension** (5-28) A (29-35) 1 Taking instructions 71.67 (12)* 70.86 (28) or helping 3 Speaking-Signalling 66.13 (367) 78.48 (511) 4 Persuading 70.59 (76) 78.12 (119) 9 Monitoring 65.34 (29) 78.10 (21) * Sample sizes are shown in the parenthesis ** Only.the levels shown were represented in the sample. 64 range of up to 35, were split at the median number of observations yielding two levels of source credibility; low credibility (5 to 28) and high credibility (29 to 35). Examination of the mean values in the table reveals that the mean perceived probability of occurrence of expectation scores are fairly close across levels of source credibility for the lowest level of the PeOple job dimension. However, in most of the levels of this job dimension, the strength of expectation ratings for participants who rated their sources of information as being highly credible are stronger than the ratings of participants who perceived their sources as lower in credibility. Those participants with lower source credibility ratings actually show a small drop in strength of expectation ratings for low to high levels of dealing with people on the job. This is in contrast to the rise in perceived probability of occurrence of expectation ratings which occurs from the lowest level of the People dimension for participants who have highly credible sources of information. This difference in means suggests that, particularly for jobs at middle and upper levels of the people worker function (e.g., engineering, computer programming, accounting/finance), the credibility of the sources of expectation information may be important in determining the strength of the participants' expectations. The Interview Sample. The results of the regression analysis for the interview sample are shown in Table 13. q A For the main effect variables, R = .089. For the main Table 13 Regression Analysis Summary Table for the Interview Sample Dependent 2 Variable Step Variable Entered F to Enter R AyyOverall F Probability 1 Source Credibility (C) 16.03 .055 of Occurrence of the People (P) N.S.* N.S. Expectation Data (D) 6.81 .089 Job Search Position (8) N.S. N.S. Things (T) N.S. N.S. 5.56 (df = 5/251, . p< .05) 2 ‘C X T 8.37 .146 C X S N.S. N.S. C X P 4.28 .169 C X D N.S. N.S. 5.63 (df 8 9/247, p< .05) *N.S. = Not significant 66 effects-plus—interaction terms model R2 was equal to .170. A test of the difference between the R2 values was signifi— cant (F = 5.35, d.f. = 4/247, p( .05), indicating that the interactions contributed significantly to the perceived probability of occurrence of the expectation. Examination of the F ratios for the regression coefficients shows that, of the main effect variables, Source Credibility and the Data job dimension are significant contributors to the equation. b As in the questionnaire sample, the main effect of Source Credibility accounted for a large portion of the variance in the dependent variable (Probability of Occurrence). This is also seen, as it was in the questionnaire sample, in the significant simple correlation between source credibility and the strength of the expectation (r = .23) (Table 14). The other significant main effect variable in the regression equation for the interview sample was the Data job dimension. Table 15 provides the mean perceived probability of occurrence of expectation scores for the levels of the Data dimension that were represented by the prospective jobs of the participants in this sample. The mean scores for participants who were searching for jobs involving the lower levels of dealing with data (e.g., sales, personnel) were higher than the expectation ratings of participants who planned on taking jobs that dealt with data at the more advanced levels (e.g., engineering, 67 Table 14 Correlation Matrix of Variables in The Regression Equation in the Interview Sample >. 3:8 >. :15 mi: o = ' m 8: 8: 3 a g 63 ,Q: 3;: (3 o .QHJJ U: D 04 0:) CW! 2 'm I; 016*) x x x HO may a, E4 (to)!!! U U o no a; . :00 ‘H H a. to 0 Variable Source Credibility 23* Data -14 -02 People -10 01 -18 Things -00 —07 42* -17 Job Search - Position 10 -11 -04 -27* 18 Source Cred X*** Search (CXS) 22* 36* -05 -29 15* 87 Source Cred X Data (CXD) 05 70* 69* -11 24* -11 22* Source Cred X People (CXP) 03 38* —17 91* —18 -32* -15 15 Source Cred. X Things (CXT) 00 12 40* -17 96* 17 22* 38* -11 * Significant at p<.05 ** Dependent variable *** Source cred - Source credibility 68 Table 15 Mean Perceived Probability of Occurrence of Expectation Scores for Levels of the Data Job Dimension in the Interview Sample Data Job* Standard Number of Dimension Q Mean Deviation Observations 4 Compiling 88.49 13.64 53 5 Analyzing 88.57 8.52 7 6 Coordinating 79.80 20.46 150 7 Synthesizing 82.12 16.92 85 * Only the levels shown were represented in the sample. 69 accounting/finance). This finding in the interview sample is somewhat in contrast to the findings of the questionnaire sample, in which engineers, as examples of the high levels of the Things dimension, tended to have stronger job expectations. The same relationship between the Things job dimension and the perceived probability of occurrence of the participant's job dimension would have been anticipated in this sample, too, since the Things and Data job dimensions were correlated significantly (r = .48) (Table 14). However, in the interview sample, the main effect of the Things job dimension was not a significant predictor, although the interaction of Source Credibility and the Things job dimension (C X T) was significant. Table 16 shows the mean strength of expectation scores for the two extreme levels of the Things job dimension which were represented in the sample. These mean scores reveal that participants who were planning on taking (or had already accepted) jobs at the upper level of dealing with things (e.g., engineers), appear to have stronger expectations about the job than did participants who were searching for other types of jobs. This is only true for participants who rate their sources as lower in credibility. For participants who perceived their sources to be highly credible, the difference in perceived probability of occurrence of expectation between the low and high levels of the Things job dimension was in the opposite direction. 70 Table 16 Mean Perceived Probability of Occurrence of Expectation Scores for the Interaction of Source Credibility X Things Job Dimension in the Interview Sample Source Credibility low high Things Job Dimension** (5-28) (29-35) 1 Handling 80.66 (98)* 84.72 (116) 7 Precision Work 86.29 (21) -78.50 (20) * Sample sizes are shown in parenthesis ** Only the levels shown were represented in the sample. 71 Table 17 Mean Perceived Probability of Occurrence of Expectation Scores for the Interaction of Source Credibility X People Job Dimension in the Interview Sample Source Credibility low high People Job Dimension** (5—28) (29-35) 1 Taking instructions 0 (0)* 90.00 (6) Helping 3 Speaking-Signalling 80.96 (88) 81.78 (86) 4 Persuading 86.72 (29) 88.87 (31) 9 Monitoring 64.00 (5) 82.3 (13) * Sample sizes are shown in parenthesis ** Only the levels shown were represented in the sample. 72 There is, however, a large discrepancy in the number of cases in this relationship, with the number of observations in the upper level of the Things dimension being much smaller than in the lower level. This may cast some doubt as to the interpretability of this interaction and, also, to the significant interaction between Source Credibility and the People job dimension.1 The interaction of Source Credibility X People job dimension (C X P) is shown in Table 17. As in the questionnaire sample, the mean strength of expectation ratings were higher for participants who perceived their sources as being highly credible, although, in this sample, the difference is rather small. The small number of cases at the lowest and highest levels of the job dimension also hinder the interpretability of this interaction. Summary. In both the questionnaire and the interview samples, the credibility of the first listed source of expectation information was found to be the strongest contributor to the perceived probability of occurrence (or strength) of expectation. The type of job for which the participant had applied (or accepted) also had some , influence in the perceived strength of expectation, with applicants for engineering position (represented by high "Things" and moderate to low ”People" job dimension ratings) having stronger expectations, particularly in the questionnaire sample. 'The questionnaire sample also showed some relationship 73 between the interaction of source credibility and the ”People" job dimension (C X P) and the dependent variable, although this was not very strong in either sample. The job search position variable was not shown to contribute significantly to the perceived probability of occurrence of expectation. However, further analyses, unrelated to the research propositions, did reveal a trend in the overall certainty of the participants' expectations (as measured on the seven-point semantic differential scale with ”Very Certain“ and "Not Certain at All” as end points) which was moderated by their position in a job search. Participants who had already accepted a job were more certain of their expectations in general, than those who were searching for a job but had not yet accepted a position. This was true of participants in both the questionnaire sample (i pre-accept = 4.90; i post-accept = 5.55; t(98) = -2.24, p<.05) and the interview sample (i pre-accept = 5.20; i post-accept = 5.72; t(66) = -2.17, p<.05). Proposition IIIa Thirdly, it was proposed that expectations derived from interviews with job applicants would be perceived as having a higher probability of occurrence and would be more strongly related to anticipated job satisfaction than those ratings of expectations which were presented to applicants from a pre-determined list. Examination of the applicants' perceived probability of 74 occurrence of their expectations on the job revealed that for the questionnaire sample the average estimate of the perceived probability of occurrence of the expectation (i = .72) was significantly lower than for the interview sample (3": = .82), (t = -7.o7, p<.05). A Pearson product-moment correlation was performed in order to assess the relationship between the perceived probability of occurrence of the participant's expectations (and their evaluative components) and anticipated job satisfaction. Separate correlations were computed for the questionnaire and the interview samples, with a significant relationship being found in the questionnaire sample (r = .35, p( .05) . The correlation in the interview sample was smaller and not significantly different from zero (r - .18, p) .05) . A test of the significance of the difference between the two correlations was performed after an r to z transformation. The resulting z score of the difference between the correlations was not significant (2 = .42, p) .05) . This suggests that, contrary to the proposition, the ratings of perceived probability of occurrence of expectations from the interviews did not correlate more strongly with anticipated job satisfaction than did the ratings expectations from the questionnaire sample. Proposition IIIb It was further proposed that job applicants' expectations derived from interviews would be different in content from those expectations contained in the 75 pre-determined list used in the questionnaire method. To test this proposition, expectations elicited from the interview sample were categorized and compared with the expectations in the questionnaire list. Table 18 contains the ten most frequently cited expectations, listed according to their frequency of occurrence in the total number of expectations generated from the interviews (N=308). The table also includes additional sample information for each expectation: the average evaluations score (from the seven-point Good/Bad semantic differential scale), the average probability of occurrence rating of the expectation on the job (from zero to 100%), the percentage of male and female participants who cited the expectation, the percentage of participants who currently were still involved in a job search (pre-accept) and those who had already accepted a position (post-accept), the percentage of participants who planned on taking jobs in engineering (Eng.), business (Bus., including accounting, sales, marketing, etc.) or other areas, and the average rating of-anticipated job satisfaction of the participants who cited that expectation. Average and total ratings for the entire interview and questionnaire samples are also listed for the purpose of comparison. Half of the expectation categories in this list were not represented in the pre-determined list of expectations used in the questionnaire. This indicates that job applicants may have, as part of their belief systems, 7’6 n~._ 00.00 no. nooeu :30 so: ee.. n_ n. ex en «a as _n ee.n~ n..e c on nueteeneeee oosomuoaxu «e. e_.~ n«._ neeenuua\neeeenna n_.n nu AN 00 0a 00 n0 nn n0.00 «0.0 0 o>«u«oom 0cos< uoo cu ammu. ee.n_ ee.s nueeeueeeee no.0 s s 00 00 00 00 0N 00.x0 00.0 n o>¢= or 00. «n.0— 00. n_.n as we . no e. on as an ne.on no.0 n queuesotu hm. -.s— 00._ casvoeumxvsoa 00.0 0n 0 0n _n 00 00 N— 00.00 s0.n n .xuos noon huuuunquconuoz mm. 0~.n~ 00. we n~.m en nu an eN e. me me ne.eu e..o e autos and: co. n~.n. so. seesaw 00.0 an 0 en A“ no an 0— 0—.00 00.0 N0 0000 occuuaaoz en. an e. _e an as on .N ea.e_ Ne. seconnoetusee 0—.0 00.00 50.0 n— 0000 .e.u\m u u .u a e» an en: .a.m\m .a.m\m en eoeeunuoexn new 000 quSuO .ucn u.osn uaouuu oosbuusuuc 0>< on new season 000 up .0I0oum we .vouh assay-uoonxu townhousu no addendu000qugu vs. nouucoauuum 0— oualh 7'7 coauouu-«unn 00n oncuo>< I .uum 060 .u>< Annunnxoun .uooulono .unuuuuoouov case undue-suns» I .usn O needy-don mean“ 0 .mcuuuuuou .aossoouoa .mcuuuxuul \0suouuuopvo .uonc. .uaolouacgl .ousaauu .ucuussououv aaon encodes: I .090- .so«uouuonno so.» candy 0:! duals. souvuousu nous» no nussvubuvsu no unsucoouomc en. .Nn.nn ea. «Hanan ee.n en e_ es ~n no on N. nn.~e na.o use... one-eeoeeauea “such es. e~.e_ a .3 cent: use-um __.n an e_ an AN we no on .n.~o ea.» nuetnonee sauce sadne~0e< eta ea. ae.n_ no. needs» etc on: o_.n on oN on em as a. an en.ee «a.» c on neeetneoeeo es. oe.n~ ~._ gauge «n.. a .N as on es on so ~a.ee ce.n . son assuage» .35.. u u a u a H .8. .eém 5.3m w, 8:383 0.0 mon unusuo .usn c.000 unuuucluuom unsou0 ouuouuaouc n>< dams 000 :uusoa 000 no .aoaoua uo .vouh AvoscuusoUv 0— odaah 78 expectations that are not being assessed in existing measures of job expectations. For example, although most pre-determined lists of expectations are overwhelmingly positive in their orientation, this list of interview- elicited expectations shows that some expectations of applicants may be less than positive. The expectation category of 'Poor Work Load Schedule" has a much lower average evaluative rating (i z 3.87) than the other more ”positive” expectations of interview participants (i = 6.45), a difference which is significant (t p(.01) The probability of occurrence of each expectation in (292) = 12.70,_ the interview sample does decline slightly as its frequency of citation becomes less. However, as shown earlier in the test of proposition IIIa, when compared with the question— naire sample, this measure of strength of expectation is still significantly higher in the interview sample. For the most part, the characteristics of the participants who cited each expectation (including sex, position in the job search, and the type of job being sought) reflected the general distribution of these characteristics in both the interview and questionnaire samples. One exception is for the expectation of ”Specific Job Tasks.“ This expectation category, unlike the rest of the sample, has a higher percentage of males than females citing such expectations, as well as a higher percentage of participants who had already accepted a job. Since this 79 category also has one of the higher percentages of prospective engineers, it seems reasonable to assume that those participants who had already accepted a job were most often male engineering students and that they had some of the more job-specific expectations of the interview sample. 'The anticipated job satisfaction ratings do not reflect any pattern in the statement of expectations by the participants in the interview sample. However, the average anticipated job satisfaction score for the entire interview sample (i = 5.11) was lower than that of the total questionnaire sample (i = 5.74). This difference was significant (t(178)=6.03, p.<.01) and could possibly be the result of the inclusion of negative expectations by some of the interview participants, as opposed to the all-positive expectations rated in the questionnaire sample. Summary The results of this study have provided support for the proposition that job applicants recognize various sources of information for their initial (pre-job) expectations. Although sources close to the organization (i.e., recruiter/ interviewer, employee of the organization, personal exper- . ience on a job) predominate, other sources that are not related to the job or organization (i.e., teachers, family, friends) were also cited, particularly when more than one source was listed. The credibility of the first listed source was also found to contribute to the strength of the expectation (as 80 measured by the perceived probability of occurrence of the expectation) in both the questionnaire and the interview sample. In the questionnaire sample, small, but significant contributions were made to the strength of the expectation by various job dimensions (People and Things) and by the interaction between Source Credibility and People (C X P) dimensions of jobs to the strength of expectation. The strength of the job expectations was further shown to be greater (i.e., given a higher probability of occur- rence value) in the interview sample, in which participants were allowed to state their job expectations, than in the questionnaire sample, where participants rated the strength of expectations from a pre-determined list. Contrary to Proposition IIIa, though, the expectations from the interviews (along with their evaluative components) did not correlate better with anticipated job satisfaction than did the expectations from the questionnaire sample. However, the correlations within both samples were low and the difference between the correlations was not significant. This suggests that neither the expectations from the interview nor the questionnaire samples may be strongly related to an anticipated attitude about the job. The content of the expectations elicited through interviews with job applicants did include some of the expectations from the pre-determined list used in the questionnaire. In addition to these, interview participants expressed positive expectations of good interpersonal 81 relations on the job, positive training and learning experiences and challenge. Participants also anticipated some negative occurrences on the job, such as heavy work loads and long hours. CHAPTER IV DISCUSSION The purpose of this study was to determine the content and sources of applicants' expectations about their next jobs. The contribution of the credibility of these sources to the strength of the individual's expectation and the relationship between the strength of expectation and anticipated job satisfaction were also assessed. Applicant Expectations About the Job The content of applicant expectations was most clearly reflected in the expectations that were elicited from interviews with participants. The most frequently cited were positive expectations about the job. These included a mix of statements that represented both the expectation categories from the pre-determined (questionnaire) list and the newer categories developed from this study. These expectations reflected the current concerns of individuals soon to be graduating from college and entering the job market. For example, good salary was one of the most often cited expectations. Participants also expected the job to make use of the skills, abilities, and ideas which they had developed in school. However, there were also expectations of training programs that would provide them.with more experience as well as help in adjusting to a new job and environment. The interview participants also expressed expectations for good interpersonal relationships on the job with 83 customers and co-workers. This suggests that participants expect to encounter peoplewith whom they will be able to work well. Statements in this category included expectations for co-workers who would be friendly and also helpful to the participant in his/her adjustment to and work on the job. These positive expectations about the job may also be seen as anticipation of a pleasant and almost “ideal" working atmosphere, in which participants would be given the opportunity to be "challenged” to use their skills, but would be able to turn to training programs and co-workers for assistance if necessary. Negative expectations were also listed in interviews with job applicants. Participants frequently expressed . expectations of long hours and heavy work loads, partic- ularly when beginning a new job. Although these expecta- tions were not pleasant or positive (receiving significantly lower evaluative ratings than the other more positive expectations), they were usually expressed without any sense of dread on the part of the participant. These often-cited expectations, both positive and negative, were distinguished by their level of generality. The majority of the expectations listed by the interview participants were fairly general in their content and could be applied to almost any type of job. These rather general (i.e., non-job—specific) expectations are particularly interesting coming from a sample which consisted of 84 individuals who were searching for very different types of jobs (e.g., engineering versus business management) and who were at various levels of their job search, from those who were just beginning to look for a job to individuals who had accepted a job offer. Despite these variations in the job type and job search levels, and, consequently, the amount of job-specific information which the participant could have received, the expectations remained general in their content for most of the sample, with very few being considered ”job specific." This level of generality in expectations may indicate that these are not so much specific expectations, derived from informed sources, as they are desires of the applicant. These desires may have been part of the individual's concept, of a job (or of work in general) from an even earlier (i.e., pre-job Search) point in their vocational development. Although it was carefully pointed out to the partic- ipants in both the questionnaire and the interview methods that they were to rate what they expected to occur on the job they planned to take, participants may have listed and rated their hopes or desires for the job. Bray et al., (1974) have found the correlation between desires and expectations to be high (r = .87), indicating the optimism of new employees who felt that what they hope for is what will occur on the job. The expectations as desires expectation may also account for the primarily positive nature of the expectations expressed by this sample because 85 an individual's hopes and desires for a new experience are seldom negative, especially at the start of the job search process. The description of these stated expectations as "desires“ is not to say that desires are not important and influential in affecting the applicant's initial (and later) job attitudes and behaviors. These desires, are expecta- tions, or beliefs about the new job situation, and as beliefs, they have their basis in information from sources, either self-experience or others. Source of Expectation Information Examination of the sources of expectations in both the questionnaire and the interview samples suggested groupings of organizationally-related and non-organizational sources. The organizational sources were found to be more often listed as the first source of expectation information. This reflects the assumption made by other expectation researchers (Bray et al., 1974; Dunnette et al., 1973; Ward & Athos, 1972) that sources close to the organization, such as recruiters, are the primary sources of job information. However, when more than one source of information is listed by the participant, non-organizational sources were found to be listed just as often as organizational sources of expectation information about the job. Sources of infor- mation, such as school and family do seem to leave an impression on the job applicant, even after the contribution of information from sources closer to the job or 86 organization has been acknowledged. Source Credibility and Expectations The impressions left by sources of expectation information did not appear to be limited only to the recall of the identity of these sources. The credibility (or believability) of the source (specifically the first listed source) was found to contribute to the perceived probability of occurrence (or strength) of the job expectation in a regression equation. In both the questionnaire and the interview samples, participants who perceived their sources of information to be very credible appeared to have stronger expectations (i.e., be more sure of the probability of occurrence of their stated expectation) than did those participants who rated their source as less believable. This suggests that the information provided by these sources does seem to have some influence on the job expectations of applicants. Along with source credibility, the analysis of the questionnaire sample revealed small but significant contri- butions made by certain characteristics of the participant's anticipated job type on the strength of the expectation. Specifically, stronger expectations tended to be held by participants whose prospective jobs involved dealing with things or machinery at the advanced level (i.e., setting-up versus handling) and contact with people at the level of speaking to or persuading, particularly when that level of contact with people was combined with information about the 87 job from.what was perceived as a highly credible source. In this sample, and especially in the questionnaire sub-sample, the job of engineer best fit the qualifications of being at the advanced level of the Things job dimension and at least the middle level of the People job dimension. Participants who were searching for engineering positions, especially those who perceived the sources of their expectation information as being highly credible, tended to have the strongest expectations about their next job. Expectations and Job Attitudes Although the strength of the expectations (as measured by their probability of occurrence on the job) was signi- ficantly higher for the interview participants than for those in the questionnaire sample, the data did not support the proposition that interviews would elicit expectations that were more salient to the applicant than those for pre-determined lists. The correlation between expectations and anticipated job satisfaction for the interview sample was not significant. The correlation for the questionnaire sample was low, but significant. A test of the differences between the correlations for the two samples was not significant. The low correlations in both samples do not provide much support for the role of job expectations in the formation of job attitudes. The low correlations, however, may have been due to problems with the job attitude measure. One possible problem with this measure is the use of an 88 Expectancy-Valence theory often uses anticipated attitudes and behaviors as outcome measure (c.f., Bartol, 1976; Mitchell & Knudsen, 1973), the use of anticipated job satisfaction in the present study may have been an inappropriate measure of job attitude. The fact that participants in this study were requested to rate their anticipated or future attitude about the job does not mean that what they rated was not an attitude. It may be argued that an individual who has not yet experienced a situation may not be able to form an attitude about it. However, even an expectation of a job experience can contain cognitions, behavioral intentions, and affect anticipated by the individual. This is not to say that the anticipated (expected) attitude will be the same as the actual (experienced) job attitude, but that the job satisfaction anticipated by the applicant still has the elements necessary for an attitude. Another problem with the job attitude measure was the possibility that participants were not clear about which "anticipated" job they were to rate. Although participants in both the questionnaire and the interview samples were told to rate the job that they planned to take, it is possible that they were unable to envision and assign values to a specific job, particularly if they were at an early point in their job search process. Perhaps they were rating their anticipated satisfaction of a "composite" of jobs, which could also account for the general nature of the 89 expectations. A more probable reason for the low correlations between expectations and anticipated job satisfaction is the limited variance of the unidimensional measure of job satisfaction used in the present study. Other research using the Fishbein and Ajzen formula (e.g., Fishbein & Coombs, 1974) has used multidimensional measures of the attitude object. However, in the present study, the Faces scale (Kunin, 1955) was used as a single well-established measure of job satisfaction. Despite the acceptance of such single measures of job satisfaction by researchers in the Organizational Behavior area, there has been some question as to whether they really represent attitudes about the job. Wild and Dawson (1972) have shown that there is a difference between the ratings of overall job satisfaction and attitudes towards specific aspects of the job. They believe that this difference may be mediated by other variables, such as age, and marital status of the individual. The authors caution that, “It is clear that the attitudinal basis of overall job satisfaction is not static,” (p. 157), with different attitudes being dominant in the job satisfaction of various groups of workers. The anticipated job satisfaction ratings of the participants in the present study were fairly high, with little variability in both the questionnaire and interview samples. Whether this measure was inappropriate because of 90 its anticipated nature or its unidimensionality, subsequent tests of the relationship between initial job expectations and attitudes about the job would be best made using measures of specific job attitudes, preferably taken from individuals who have been on the job for some period of time. Implications for Future Research Methodological problems in doing job expectation research are among the issues to be considered in future research on initial job expectations. Measures of the expectations of applicants should include both positive and negative options for each expectation, or if possible, a range of the degree to which each expectation will occur on the job (e.g., low salary, moderate salary, high salary), so that the applicant may be able to more accurately express his/her expectations. To insure that the most salient of these expectations are represented, the use of interviews is suggested. Categories of sources of expectation information should allow for sources that are close to the job or organization and for sources that are more removed from the job or organization. These categories should also include options for sources that fit into more than one category (e.g., a friend who is also an employee of the organization) and an effort should be made to determine when each source was thought to be an influence for the applicant. Credibility scales for these sources of job expectation information 91 would be more accurate if they were made more appropriate for the mix of person and situational sources of job information (e.g., employee of organization, previous job experience, newspaper articles). Research on the relationship between job expectations and job attitudes should use composite measures of specific job attitudes to match the dimensions of the individual's expectations about the job to insure better reliability and validity. The variability of both the expectation and the attitude measures should then more closely correspond. A more representative sample of a variety of job types or a more in-depth study of the expectations of a single job may also shed more light on the relationship between expectations and attitudes by allowing for more accurate measures, and more control over other job related variables which may affect this relationship. Further research into the area of job expectations should consider looking into the past at the early sources of influence (i.e., pre-job search) on formation of expecta- tions. Research should also look forward, into the relationship between these expectations and the on-the-job attitudes and behaviors. This suggests a longitudinal approach to job expectation research, which will be necessary to provide the developmental perspective needed, not only in this area of research, but also in Organizational Behavior research in general (Schein, 1980). 92 APPENDICES 93 Appendix A Questionnaire Format Cover Letter to Participants Job Expectations Survey Instructions Example of Questionnaire Format Request for Job and Personal Information 94 MICHIGAN STATE UNIVERSITY mm W mm m m ' muses ° ‘3‘ ”'C'M' m' mm May 10, 1982 Dear Student; Place-at Services ad the Depart-ant of Psycholog at Michigan State University are co-sponsoring a investigation of the mutations that Job applicants bring to the job ad the sources of infornation for these expectations. Past research has shown that people do come to the job with expectations about what the job will be like for then. When these expectations are disconfirned by actual enerienc. on the job, increased job dissatisfaction ad turnover may result. This study is an atteIpt to determine the content of these expectations and the sources of this intonation in order to enhace students' chance of aployeent ad facilitate their adjus‘tnent to the Job. This interaction will be collected through a questionnaire. We are interested in contacting studats who are actively searching for a per-nan: Job (i.e., who have applied or requested a intervi-r with an organisation). Studats my have already accepted a job offer, but they nay not yet have actually been on the job. Students who voltmteer to participate in this investigation will be entered in a cash drying with a prize of 5100, provided by the principal investigator. Participats will also receive a brief copy of the results of the study. which should help then better prepare for their job search and initial job experiences by providing specific intonation on the unrealistic (and realistic) Job expectations of college students. If you agree to participate in this study. please coalete the enclosed questionnaire (including your signature on the consent sheet below) ad return the entire packet in the envelope provided. Your nae will only appear on the consent fern, which will he revved front the questionnaire ad used in the cash drawing. Your responses will be kept confidential ad your participation is voluntary. If you have any questions. please contact Paula Popovich at 353-4591 or 332-8035. Thank you. Gordon Hood [Io U Professor ad Chairperson Director Depart-at of Psychology Plat-at Services I havereedtheesplanationofthepurposead (please print nae) procedure of this study ad agree to participate. I understad that Iy participation is voluntary. that sy results will be kept confidatial ad that I .y'quit at any ti. if I feel comfortable. PLEASE RETURN 3‘! JUNE 12. 1982. (Signature of Student) (Per-aat Mailing Address) M“, II on Am Ans-affine! Omen.“ W 95 JOE EXPECTATIONS SURVEY INSTRUCTIONS The purpose of this questionnaire is to find out what you expect your (next) job to be like. On each page you will be given a statement about what may occur on a‘ job. Following each of these statements are two scales consisting of two words separated by seven lines. such as GOOD BAD. Place a check on the line which best describes how you feel about each statement in gaperal (not necessarily how it will apply to your job). You will also be asked froe what source you received the information that this statement describes some- thing that would or would not be part of‘ygg; job. -£semples of possible sources of information about the job are listed below. Please feel free to use any sources. even those that are not included on the list. Parents Siblings Other Relatives (specify) Friends Teachers School Counselors Placement Services Recruiter/Interviewer Employee of the Organisation Organizational Advertisements Personal Experience on the Job Experience With the Job as a Customer/Consumer Television Radio Newspapers Magazines Other Sources (specify) You may list the sac source for more than one job statement, but please rate them separately each time. If you have difficulty rating the sources on these scales. give your first impression. Remember. these responses are your opinions. there are no right or wrong answers to these questions or ratings. Your responses will be anonymous and will be kept confidential. Please make sure you answer_I11_parts of the questionnaire. If you have any questions. please contact Paula Popovich at 353-4591 or 332-8035. Thank you. 96 GOOD SALARY A. GOOD BAD a. mat ______' __IENS?ICIAL c. mt umcnmcrruiruoooo mrmmnpm orrounaonz (0-1002) 4 2 Where did you get the information that a good salary would or would not be part of your Job? Use the list of sources on the first page or any others. write the name of the source in a space below and rate each source on the scales Shit follow. If you have no source for this information, check the "No Source" space below. SOURCE“ mum ________ummmtr m__________ruasm WISH __________UHSEL!’ISH m _________________AHPUL mm _____________m was: ______________IUTATALL W10!!! WIDE Howachof whatyouhnowaboot this aspect of the Jobwas providedbythis source? (0-1002) 2 SOURCE” mu _____________Ulnnlm.! m_______rmsan scum ___________mlsn.nsn not ______________AHPU‘L an!!! __________m m! _______________IIUTATALL WIDE: mute lowacbofwhatyouhnowabout thisaspectof thejobwasprovidedbythissource? (o-xooz) ' : ADDITIONAL ms m m 5. AUMQM 10. 11. F 97 In this questionnaire, you have described what you expect your Job to be like. How certain are you that what you have described will actually occur? VERY CERTAIN __ __ __ __ __ __ __ NOT CERTAIN AT ALL How far along are you in your search for a job? (Check the statement that best describes your current situation.) _a) I am not currently looking for e) I have applied/interviewed for a Job more than one Job __b) I am thinking of looking for a Job _f) I have had a Job offered to u _c) I know of at least one Job that I (but h." not fit accepted it) am planning to apply for g) I have accepted a Job offer _d) I have applied/interviewed for h) Other (please describe) at least one job Have you talked to other students about looklng for a Job? What have you discussed? Have yourexpectations about what the Job will be like begun to change since the beginning of your search for a Job? How? Once you are on the Job. how satisfied do you think that you will be with that job? Circle the face that best describes how you feel. (As-308.10 How would you react if what you expect the Job to be like is not what the Job is actually like? How satisfied would you be in the job if it is not actually what you expect it to be like? when on some Are there any other things that you expect to be part of your Job that are not covered in this questionnaire? what are they? that type of Job are you applying for? How many Jobs have you had before this one? What is the longest aaount of time you have spent in any .o_I_Ie_ Job? anths How lacrtant is your next Job to you? _ VERY IMPORTANT __________ MOT IMPORTANT AT ALL Is this Job: PERMANENT or TEMPORARY SEX ID (thefilast four diifts of your social security number) 98 Appendix B Interview Format Cover Letter to Participants Interview Format (Instructions to Interviewer) Interview Reporting Sheet 99 MICHIGAN STATE UNIVERSITY mm or monomer WT um ° W ' .30 "'°'°‘°°" M' ”M" May 10. 1932 Dear Student; Placaant Services ad the Depart-at of Psycholon at Richiea State University are co-sponsoring an investigation of the expectations that job applicants bring to the Job ad the sources of information for these expectations. Past research has shown that people do cone to the Job with expectations about abet the job will be like for than. than these expectations are disconfirmed by actual experiences on the job, increased job dissatisfaction ad turnover lay result. This study is an attapt to detaruine the content of these expectations ad the sources of this infornation in order to enhace studats' chance of enployaat ad facilitate their adjust-ant to the job. This intonation will be collected throueb a brief inter-vie! ad a follov-up questionnaire. Re are interested in contacting students rho are actively searching for a paraaent job (i.e., rho have applied or requested _a inter-via with an or'aisation). Students ay have already accepted a job offer, but they my not yet have actually been on the job. Students rho volunteer to participate in this investieation rill be atered in a cash drawing with a price of $100. provided by the principal investigator. Participate will also receive a brief copy of the results of the study. ehich should help tha better prepare for their job search ad initial job experiaces by providing specific inforsation on the unrealistic (ad realistic) Job expectation of college students. lie will be contacting students by telephone for a short (i.e., 20 ninute) inter-via about ehat they expect their future job to be like. If you agree to participate, you can wait until you are contacted, or you can call Paula Popovich at 353-4591 or 332-8035. Your responses will be luapt aonynous ad participation is voluntary. Thank you. ; E "' 3;;— '8”;- ‘y' . Gordon Hood hn Professor ad Chairperson Director Deparuent of Psychology Plan-ant Services I havereadtheexplaationofthepurpoeead (please print nae) procedure of this study ad agree to participate. I understand that ey participation is voluntary. that ny results will be kept confidential ad that 1 any quit at ay tine if I feel incafortabla. (Signature of Sttdat) mt e as AMI—saw: Anise/Eyes! (prov—mm 100 INTERVIEW FORMAT 1. Identify yourself. "Hello, I'm from Job Expectations Project. We sent you a letter informing you that we would be calling you to get some information about what you expect your job to be like. Did you receive the letter? Ask permission to give ”If you are willing to the interview. participate in this study and will sign the consent form at the bottom of the letter, we can have a brief interview now if you have the time.” Conduct the interview. - Make sure that you get their permanent mailing address so that we can send the results to them (and the prize if they win!) Tell the student that they will be getting a follow-up questionnaire to be filled out at soon as possible.- - Make sure that we have their correct current address to send the questionnaire to them. **** - Tell the student to send back the questionnaire, with their signed consent form Thank the student. Ask them if they have any questions. If they do, answer them as best as you can and tell them to contact me if they have any other questions. Call-in the names and expectations to me (353-4591 or 332-8035) as soon as possible, and I will send out the questionnaire. 101 INTERVIEW PERMANENT NAME ADDRESS 1. What sorts of things do you expect (believe) will be part of your job? Please generate a list of things that you expect to be part of your job. What will be important to you on your job? (not necessarily all positive) EXAMPLES (use only if desperate and be general) Working conditions Co-workers The work Describe how your life on the job will be different from your life now from 8:00 a.m. to 5:00 p.m. 2. Specific expectations/beliefs (Read back to interviewee) 3. Is there anything else? 4. If there are expectations, list them on a questionnaire and let the interviewee know that they are being send a questionnaire to fill out as soon as possible. 5. If there are ng_expectations, fill out the brief information sheet while you have them on the phone. 6. REMEMBER: We can use the individual if they are looking for a job or have even accepted the job. We cannot use anyone who has already been working on the fiob that they are telling us about. --If you reach someone who has accepted a job offer, please tell them that the Placement Center needs that information for their files. Have the student call the Placement Center. ' moun> pucoomm u m mcfiuom cofiuocom noxuo: accumwuo u o « magmam: Icoauonuumcn mcaxoa a m adduced: fl 5 mou>num N h mswuounHMOImcwpuom N m moaadnco«wlm:«xmomm m w msauomeou H w mcwpcma m m maupobmuwm w m adammoo N m ocuunaomwcoz v v mcfiuuubwa m w unauomeou m v mcwumuomolmcw>quo m m mnwmd>uua5m w m mafiadmeou v m M ocuaaouu:OU|mcaunquo o m ucwuoouumca b N unashamcd m N 1 upon :ofimoouum a H unannouoouz a i mesomaanuooo e H a: manuuum a o mcnuoocuz a o unannounucsm a o m D m 0 am «0 Huang moam> Huang u=Hn> dunno usflm> mmmmmm mmmmmm mwmm monan> cocooum Hausa one mao>uq noduocoh nuxuox no maunoq U xdpcmmud Worker Function Ratings for Job Types Represented in the Main Study Sample 103 Code Job Type Data People Things 01 Accounting/ Finance 1 6 7 02 Engineering 0 6 1 03 Management 1 6 7 04 Sales 3 5 7 05 Advertising/ Marketing 0 6 7 06 Computer Programmer/ Analyst 1 6 7 07 Personnel 2 6 7 08 Retailing (Buyer) 1 5 7 09 Social Services 1 0 7 10 Communication/ Public Relations 0 6 7 11 Research 2 6 7 12 Packaging 1 8 7 14 Food Services 3 6 1 15 Labor Relations 1 6 7 REFERENCES REFERENCES Aronson, E. & Carlsmith, J. M. Performance expectancy as a determinant of actual performance. Journal of Abnormal and Social Psychology, 1962, 65, 178-182. Bartol, K. M. Expectancy theory as a predictor of female occupational choice and attitude toward business. Academy of Management Journal, 1976, 12, 669-675. Beilin, H. The application of general developmental principles to the vocational area. Journal of Counseling Psychology, 1955, 3, 53-57. Bem, D. J. Beliefs, Attitudes and Human Affairs. Belmont, Calif.: Brooks/Cole Publishing Co., 1970. Berlo, D. K., Lemert, J. B., & Mertz, R. J. Dimensions for Evaluating the acceptability of message sources. Unpublished manuscript, Michigan State Univ., East Lansing, MI, 1968. Bray, D. W., Campbell, R. J., a Grant, D. L. Formative Years in Business. New York: John Wiley and Sons, 1974. Bruner, J. S. On going beyond the information given. In H. E. Gruber, K. R. Hammond, and R. Jessor, (eds.), Contemporary Approaches to Cognition. Cambridge, Mass.: Harvard University Press, 1957, 41-69. 106 Campbell, J. R. & Pritchard, A. D. Motivation theory in Industrial and Organizational psychology. In M. Dunnette (ed.), Handbook of Industrial and Organizational Psychology. Chicago: Rand McNally College Publishing Co., 1976. Carlsmith, J. M., & Aronson, E. Affectual consequences of the disconfirmation of expectations. American Psychologist, 1961, lg, 437. (Abstract) Crites, J. O. VoCational Psychology. New York: McGraw-Hill Book Co., 1969. Dunham, R. B. & Hermann, J. B. Development of a female faces scale for measuring job satisfaction. Journal of Applied Psychology, 1975, fig, 629-631. Dunnette,.M. D., Arvey, R. D., & Banas, P. A. Why do they leave? Personnel, 1973, g9, 25-39. Dysinger, W. S. Maturation and vocational guidance. Occupations, 1950, 32, 198-201. Edwards, W. Dynamic decision theory and probabilistic information processing. Human Factors, 1962, 5, 59-73. Festinger, L. A Theory of Cognitive Dissonance. Evanston, Illinois: Row-Peterson, 1957. Fishbein, M. An investigation of the relationship between beliefs about an object and the attitude toward that object. Human Relations, 1963, 19, 233-240. Fishbein, M. & Ajzen, I. Attitudes and Opinions, Annual Review of Psychology, 1972, 488-543. 107 Fishbein, M. & Ajzen, I. Belief! AttitudeL Intention and Behavior. Reading, Mass.: Addison-Wesley Publishing Co., 1975. Fishbein, M. & Coombs, F. S. Basis for decision; An attitudinal analysis of voting behavior. Journal of Applied Social Psychology, 1974, 5, 95-124. Fishbein, M. 8 Raven, B. N. The AB scales: An Operational definition of belief and attitude. Human Relations, 1962, 1;, 35-44. Fisher, C. D., Ilgen, D. R., & Hoyer, W. D. Source Credibility, information favorability and job offer acceptance. Academy of Management Journal, 1979, 23, 94-103. Ginzberg, E., Ginsberg, S. W., Axelrad, S., & Herma, J. L. Occupational Choice. New York: Columbia University Press, 1951. Hovland, C. I., Janis, I. L., & Kelly, H. H. Communication and Persuasion. New Haven, Conn.: Yale University Press, 1954. Kerlinger, F. N. & Pedhazur, E. J. Multiple Regression in Behavioral Research. New York: Holt, Rinehart and Winston, Inc., 1973. Lawler, E. E. Pay and Organizational Effectiveness. New York: McGraw-Hill Book Co., 1971. Lawler, E. E. Motivation in Work Organizations. Monterey, Calif.: Brooks/Cole Publishing Co., 1973. 108 Locke, E. A. What is job satisfaction? Qgganizational Behavior and Human Performance, 1969, 3, 309-336. McCroskey, J. C. Scales for the measurement of ethos. Speech Monographs, 1966, gg, 65-72. McCroskey, J. C. The effects of evidence as an inhibitor of counter-persuasion. Speech Monographs, 1970, g1, 188-198. McGuire, W. J. The nature of attitudes and attitude change. In G. Lindzey & E. Aronson (Eds.), The Handbook of Social Psychology, (2nd edition). Reading, Mass.: Addison-Wesley, 1969, pp. 136-314. Mitchell, T. R. & Knudsen, B. W. Instrumentality theory predictions of students' attitudes towards business and their choice of business as an occupation. Academy of Management Journal, 1974, lg, 41-52. Parnes, H. S.' Labor force and labor markets. In A Review of Industrial Relations Research, (vol. 1). Madison, Wisc.: University of Wisconsin, Industrial Relations Research Association, 1970. Popovich, P. M. & Wanous, J. P. The realistic job preview as a persuasive communication. Academy of Management Review, 1982, 1, 570-578. Porter, L. W. & Steers, E. E. Organizational, work, and personal factors in employee turnover and absenteeism. Psychological Bulletin, 1973, 82, 151-176. Porter, L. W., Lawler, E. E., 8 Hackman, J. R. Behavior in Organizations. New York: McGraw-Hill, Inc., 1975. 109 Roe, A. Early determinants of vocational choice. Journal of Counseling Psycholggy, 1957, 3, 212-217. Rynes, S. L., Henneman, H. G., & Schwab, D. P. Individual reactions to organizational recruiting: A review. Personnel Psychology, 1980, 33, 529-542. Schein, E. H. Qgganizational Psychology (3rd edition). Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1980. I Schmitt, N. & Coyle, B. Applicant decisions in the employment interview. Journal of Applied Psychology, 1976, 3;, 184-193. Seggar, J. F. & Wheeler, P. World of work on TV: Ethnic and sex representation in TV drama. Journal of Broadcasting, 1973, 11, 201-214. Sbrenson, J. E., Rhode, J. G., & Lawler, E. E. The' generation gap in public accounting. Journal of Accountangy, 1973, gsg, 42-50. Super, D. E. A theory of vocational development. American Psychologist. 1953, 8, 185-190. Super, D. E. The dimension and meaSurement of vocational maturity. Teachers College Record, 1955, s1, 151-163. Super, D. E. & Bohn, M. J. Occupational Psychology. Belmont, Calif.: Wadsworth, 1970. Tolman, E. C. Purposive Behavior in Animals and Men. New York: Appleton-Century-Crofts, 1932. Triandis, H. C. Attitude and Attitude Changg. New York: John Wiley and Sons, Inc., 1971. 110 An Outline of a Cognitive Model. In R. M. Vroom, V. H. Steers and L. W. Porter (Eds.), Motivation and Work Behavior. New York: McGraw-Hill Book Co., 1975, Work and Motivation. New York: John Wiley Vroom, V. H. and Sons, 1964. Organizational Entry. Reading, Mass: Wanous, J. P. Addison-Wesley, Publishing Co., 1980. Ward, L. B. and Athos, A. G. Student Expectations of Corporate Life; Implications for Managerial Recruitigg. Boston, Mass.: Harvard University Graduate School of Business Administration, Division of Research, 1972. Wild, R. and Dawson, J. A. The relationship of specific job attitudes with overall job satisfaction and the influence of biographical variables. Journal of Management Studies, 1972, 2, 150-157. ”'TITT'IMHILTIMJILI)!Mlllfilllt'iflflflflflifl)“