Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. INDUCTION AS A METHOD OF ORGANISATIONAL SOCIALISATION A THESIS PRESENTED IN PARTICAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS OF ARTS IN PSYCHOLOGY AT }~SSEY UNIVERSITY. -·· Bruce James Harvey 1981 ABSTRACT Induction, Orientation, and training have been primarily concerned with the dissemination of information on work related role behaviour. In this study a video induction programme was developed to contain information or the total role behaviour and qualify it as a method of socialization. The theory of work adjustment (Lofquist and Dawis, 1969) was used to identify dep endent variables to evaluate an induction as a method of organizational socialization. It was hypothesis ed that a socially based induction programme would increase respondent's measures of satisfaction, satisfactoryness (performance) and length of tenure. The socially based induction programme was developed and tested in a specialized plastic manufacturing company. Twenty eight new employees were assigned to control and experimental groups by their appointment dates (18 and 10 respectively) and two measures of job satisfaction were taken three and thirteen working days after their appointments using the job descriptive index, (J.D.I.) a standard check­ list measure of job satisfaction with five different job facets. Performance measures of production to stock were obtained and thes e coincided with the second measure of job satisfaction. The number of subjects who ceased employment within 90 working days after their appointments was also ascert~ined. The experimental group received the socially based induction video tape after the first measure of job satisfaction. On comparison of the J .D.I. scores bet\t~een experimental and control groups, there was no significant difference bet\o~een the pre or post-test measures on the five scales. An analysis of subjects "work" and "supervision" post-test scores catagorised as high or low by pre-test scores indicated that the socially based induction may have sensitized low pre-test scoring subjects in a negative direction on the post-test. No significant difference was found between experimental and control groups on length of tenure and recorded performance- Anxiety is postulated as a moderater variable of the J.D.I.satisfaction measures and limitations of the operationalization of the measures used i s discus s ed . The study high-lights the attrition, measurement , design, and administration problems of research in organizational settings. TABLE; OF CONTE NTS Preface List of Tables List of Figures CHAPTER I ORGAN IZATIO NAL SOCIALI ZATION 1.1 1.2 . 1.3 Introduction The Definition of Organizations The Definition of Socialization 1.4 Some Structural Properties Influencing Socialization 1 2 2 3 1.5 The Process of Individuation 4 1.6 Role Behaviour Socialized in the Organization 5 1. 7 Hethods of Organizational Socialization 5 1.8 Training, Orientation Training 6 1.9 Relevant Research 8 Cl-i..APTER II THE ThEORY OF HOr.K ADJUSTMENT AND ORGAN IZATIO NA L SUCI A LI~ATION 2.1 Introduction 2.2 Overveiw of the Theory of Hork Adjustment 2.3 The Relationship Between Satisfaction and Performance 2.4 Hypothesis C~~PTER III KETHOD 3.1 Experimental Design 3.2 Selection of the Organization 3.3 Control Group Subjects 3.4 Experimental Group Subject .. 3.5 Setting 3.6 Job Satisfaction Measure 3.7 Performance Measure 3.8 Development of the Induction Programme 3.9 Pilot Study 10 10 12 13 1 4 1 6 1 8 1 8 1 9 22 22 23 26 3 .10 Production of the Video Induction t ape 26 3 .11 Administration of the J. D.!. 27 3.12 Administration of the Induction Video 27 3.13 Administration or the Performance Keasures 28 3.14 Administration of the Job Tenure Measures 28 3 .15 Resea rch Hypo thesis 28 C Hf.. PTER IV RESULTS 4 .1 J .D. I. Heans and Standard Devia tions 4 . 2 Control Group Attrition 4 . 3 Visual Data Analysis 4 . 4 Tests fo r ANCOVA Assumptions 4.5 Analysis of Covariance Results 4.6 Analysis of Variance 4.7 Production 4 . 8 Tenure 4 .9 Summary Ch.4 P1'ER V DI SCUSSIO H .f).PPENDIC~S Appendix I _4ppendix ~- II Analysis of the Dimensions of Organizat ion Structure The Job Descriptive Index J. D.I. Questionnaire Scoring keys and Raw Data obtained 29 29 30 3 6 4 1 42 44 45 47- 54 55 71 Appendix III Standardiza t ion of Production figure s 80 Appendix IV Appendix V The Nedical and Applied ( !lillC S Jl. PU) and Research Council Social Psychology Unit Questionnaire 81 Video Tape Script 91 Figure I Figure II Figure III Figure IV Figure V Figure VI Figure VII Li t>T OF FIGURES "The Theory of Work Adjustment" "Experimental Paradigm" "J . D.I. Scale lvork Pre-test against Post-test" "J. r .I. Scale Pay Pre-test against Post-test" "J. D.I. Scale Promotion Pre-test against Post-test" "J. D.I. Scale Supervision Pre-test against Post-test" ''J . D.I. Scale People Pre-test against Post-test" Figure VIII "Tests in the Analysis of Covariance Model" Figure IX "The High and Lovl means for the Post- PAGE 11 15 31 3 2 33 34 35 3 9 test scores on the J .D.I. work scale" 43 Figure X "The High and Low means for the Post­ test scores on the J. D.I. Supervision Scale" 43 Table I Table II Table III Table IV Table V Table VI Table VII LIST OF '!'ABLES "Results of the Analysis of the Dimensions of the Organization" "Comparison between a Random Sample of the Target Organization (n=l9) and United Kingdom Normative Data on five scales of the l1RCSAPU Questionnaire" "Heans and Standard Deviations of the Pre-test and Post-test J. n .I. Scales for the Experimental and Control Groups" "Means and Standard Deviations of the Pre-test J. D.I. Scales for the Excluded Control Group Subjects 11 "The Tests for Skewness, Kurtosis, and correlation between the absolute values of the residuals and the expected values" "Tests for significance of difference in slope and variability about the adjusted means between two groups" "Ancova Summary of the J. D.I. Pay Scale" Table VIII "Ancova Su.r:ur..ary of the J.D.I. Promotion Scale 11 ?.AGE 20 24 29 30 37 38 40 40 Table IX "Ancova Summary of the J. D.I. People Scale 11 41 Table X Table XI "The Means and Standard Deviations of Average Daily Production Figures for Units to stock for each Treatment Group" "Length of Tenure and Stated reason for the subjects in Experimental and Control Groups who ceased employment within three calender months or 90 working days" 44 PRE FACt This thesis i s the results of two personal experiences . Hhile wo r king the cas i ngs depar tment (Intestine Processing) of a freezing works I met a couple of gentlemen that got fired because their verbal response was abrasive and explicit when they were confronted by the department manager over a minor incident . Later that same year I was preparing psychological reports on patients in a mental ins t itution • . The case histories revealed that some patients had initially failed to hold jobs because of maladaptive behaviour similar to what I had observed in the freezing works. It occured to me that if my vTork mates had been told what would have happened before they behaved in the way they did they might not have said what they said . It also occured to me tha t they might not end up with clinical case histories similar to the ones I had been researching . Socializing an individual into the work organization so they remained employed appeared to be more constructive to the individual and the company than socializing an individual back into society and the work force . This thesis is concerned with induction as a method of organizational socialization. The first chapter deals with the facets of organization and the process of socialization that are relevant to an induction that the organisation can use to socialize the individual to it . The second chapter deals with the theory of work adjustment and the identification of variables that could measure the effect of an induction programme that is a method of socialization. The development of the induction programme within the selected organization is in chapter three along with the design and methodology used to evaluate its effectiveness . In writing this thesis I have endeavoured to grasp an area of psychology in which I had little previous academic back ground . I thank the Department of Psychology for allowing me to undertake res earch in the area of my interest . I would like to thank the staff and management of the organization which supported the research and met material and technical expenses that were incurred in the development of the induction programme. I would also like to thank the Psychology Department for the use of their equipment and facilities and a special thanks to Mike for his technical assistance with the video. Finally I want to gratefully acknowledge the patience, support, assistance and valuable criticism of Beryl Hesketh, my supervisor. Bruce James harvey _,- ORGANilATIONAL SOCI ~ LilfTION 1.1 Introduction hill (1967) researched the food industry and discovered that forty one to fifty eight percent of all factory starters in his sample had left within the first three months of employment . From this turnover data it could be referred that oreaniza tions may not be socializing individuals with induction programmes in a suc cessful way. If a n induction proeramme i s to be devEloped as a me thod of organi za tion socia liza tion then the salient fea tures and dynamics of org8 nization and socializa tion must be examined . !n this cha pter the definitions of i nduction are a lso surveyed in relationship to the organization and the proces s of socialization. -2- 1.2 The Definition of Organizations Authors in a number of disciplines have attempted to provide definitions of organizations. Porter and Lawler (1975) analysed the definitions proposed by authors in Sociology, Management, Political Science and Psychology and abstracted five fundamental charactertistics: 1) a composition of individuals and groups; 2) a orientation towards goals; 3) differentiated functions; 4) rational co-ordination of activities; and 5) continuity through time. These features def~ne formal organizations. Formal organizations are the broad systems in western civilisation that work on the premise of maximising e~onomic gain and productivity. Historically organizational theory has gone through two phases. Classic organizational theory focused on the structural aspects of organization and has been called the study of organizations without people. Much modern organizational theory has ignored the structural component and has in turn been referred to the study of individuals without organizations. (Bennis, cited in Landy and Trumbo 1976). Payne and Pugh (1971), have combined these two approaches and offer a t~onomy of organizations based on ratings of the structuring of activities, the concentration of authority and line control of work flow. This typology enables the classification of any one organization and its accompanying psychological environment. The classification of organizations is important as it enables accurate replication and generalization of research in organizational settings. 1.3 The Definition of Socialization In the past socialization was seen as the outcome of the individual's initial adaption to society. Socialization was thus confined to the periods of childhood and adolescence. Socialization is presently considered to be - 3- a life long process due to the dynamic nature of society. Brim (1968) proposed five major differences in the socialization of adults compared with the conventional concept of socialization. Adult socialization: 1) concentrates on overt behaviour; 2) is more concerned with syntheses of old material; 3) transforms idealism into realism; 4) has a greater concern with teaching the individual to mediate conflicting demands and 5) has a content that is more specific and well defined. Adult socialization exists according to Cogswell (1968) because childhood socialization is not a fully adequate preparation for adult life. 1.4 Some Structural Properties Influencing Socialization The process of socialization is coneerned with the acquistition of a role rather than with a specific outcome of fitting a role. Cogswell (1968) outlined some structural properties influencing socialization: novice agent relationships, settings and target roles. Novices are the individuals preparing for a role and agents are individuals or other systems assisting in this preparation. Cogswell (1968) specified a type of relation­ ship between these two where the roles of the novice and agent are well prescribed. In this type of relationship the process of -socialization begins with a working consensus for their mutual participation. The concept of •relative­ anomie' is used to describe a different type of relation­ ship where there is relativity little knowledge between either the novice or the agent regarding the target role. In addition she suggests that repitition of the socializating agent's activities may reduce the effectiveness of that agent. Thus, according to Cogswell (1968), in order to facilitate socialization the agent and novice should have clear role definitions, a clear understanding of the purpose of the socialization and the activities of the agent should not be repetitive in nature. Cogswell (1968) discusses _the structure of settings most conducive to socialization; the greater the formality of -4- the setting the greater the strength of positive and negative sanctions on the individual. In addition group socialization is more likely to lead to the development of a group identity. Structural properties of the target role also facilitate socialization. If the choice of the role is . voluntary ''affective at tacbment" may occur. The concept of "time boundaries" affects the aquisition of a role; the greater the length of the "time boundary" or formal bond to the target role, the easier the socialization to that target role. The accurate perception of a role also facilitates socialization. Cogswell's (1968) structural properties and their influence on the socialization into a role will be discussed again later in the thesis. 1.5 The Process of Individuation Individuation is defined as a person's subjective mapping of the social world. (Ziller, 1964). Shein (1971) considered that the individual cam be thought of as an integrated set of social selves organized around a basic image or concept of self. The process of the individual maintaining a self concept within the organization and the formal socializing process of the organization occur simultaneously. Bakke (1953), referred to the process of the individual imposing his image on the formal organization as the personalizing process. Schein (1971) referred to this process as inovation. The concept of individuation is similar to the concepts mentioned above but it is more appropriate because it identifies the process as being related to the individuals perception of their individuality. Individuation and socialization are processes that occur simultaneously yet theoretically they are considered opposite in effect. Schein (1971) proposed a conceptual scheme of an individual's career within an organization where both processes co-exist but at different points in time. He hypothesised that organisational socialization will occur primarily in the individual~s initial employment or -5- "passage through the hierachial and inclusion boundaries". (Schein, 1971 page 421). Individuation as a process would therefore not have a salient effect on initial organizational socialization • . Bchein' s (1971) hypothesis has not been directly tested to date but it does appear congruent with observations and past research. 1.6 Role Behaviour Socialized in the Organization Entry into an organization is considered to involve t he acquisition of a new role. (Karriot, 1974). Schein (1971) distinguished three kinds of role behaviour and their importance from the perspective of the organization: "pivotal" role be~viours that are considered essential for the individual's performance "relevant" role behaviours that are considered desirable but not absolutely necessary; and "peripheral" role behaviours that are not seen as necessary f or the individual's actual work performance. Gross (1975) distinguished between technical skills, tricks of the trade, and social skills. Caplow (1964) grouped the areas in which changes occur in organizational socialization into skills, self image, involvement and values. Technical skills (Gross, 1975) and skills (Caplow, 1964) can be classified as "pivotal" role behaviours. Tricks of the trade can be classified as "relevant" role behaviours. Social skills and self image could be considered relevant role behaviours in some organizations (e.g. Insurance). Generally, the social skills, self image, involvement and values of the individual could be classified as "peripheral" role behaviour; that is not essential for the individual's performance in the organization. 1.7 Methods of Organizational Socialization Caplow (1964) identified a number of methods of organizational socialization. Anticipatory socialization involves the individual identifying with a target role and adopting the nole behaviour of its present group members. - 6 - This form of socialization gives the group considerable influence over the individual. If the group is central to the organization then the organization has that influence as well. Selection is a process of organizational socialization in that it enables the organization to ensure the desired behaviour in individuals by only selecting individuals who exhibit that behaviour. A formal apprenticeship is an explicit or direct relationship between a new employee and a present organisa­ tional member who serves as a model for the role behaviour desired by the organization. This method of organizational socialization is relatively inefficient which restricts its usefulness in large organizations. A formal apprenticeship is restricted to situations where the presence of a model is necessary due to the diversity of behaviour required for that role (e.g. mechanics, carpenters), or the role is very unique. Trial and error socialization is a common method of organizational socialization. The novice aquires the role behaviour desired by attempting different forms of behaviour with the desired behaviour receiving reinforcement and confirmation. This form of socialization is useful from the organisational perspective. It is inexpensive and its frequent usage would indicate some success with this method. However, the time required for the individual to acquire the appropriate role behaviour may be both production and labour turnover. obtained in less time by training the desired role behaviour. expensive in terms of This could have been individual in the 1.8 Training, Orientation Training and Induction The formal organizational process of socialization or imparting role behaviour required in the organization has been labelled as training, orientation training and induction. Training can be defined as "the planned activiti~s on the part of an organization to increase the job knowledge and skills, or to modify the attitudes and social behaviour of - 7- its members in ways consistent with the goals of the organization and the requirements of the job." (Landy and Trumbo, 1976 pg. 222). Schein (1968) suggested that the focus of this training was on role behaviour. Orientation training is the organization's initial method of imparting the frame of reference for the individual's role behaviour. Generally orientation training occurs once for an individual in a particular organization. Orientation programmes are established to provide new employees with information on such matters as company organization, the history of the firm, policies and proce­ dures, pay and benefit plans, conditions of employment, safety practices, names of top executives, locations of various departments and facilities, manufacturing processes and work values r· . Miner and Miner (1977) Induction has also been viewed as an introduction to the organization's purpose, policies and practices. Smith and Wakely (1972, page 206) cite the following objectives for "orientation training", "indoctrination" and "induction"; 1) Increase the employee's knowledge of such matters as company rules, wages, insurance and products; 2) Develop the skills involved in safe working habits and quality production; and 3) Develop confidence in the company and pride in its products. Induction can be seen as the initial facilitation of the acquisition of required role behaviour in the organization. Induction appears to include both general orientation and some specific training. The objectives of past induction procedures appear to emphasise role behaviours that Schein (1971) described as "pivotal" and "relevant". If induction is to be successful as a method of socializing the individual into the desired role it should include the introduction and facilitation of the total role behaviour required in the organization. As a true method of organization socialization it would cover "pivotal", "relevant" and "peripheral" role behaviour. The general hypothesis of this thesis is that induction which includes training in "pivotal" "relevant" and 11 peripheral 11 role behaviours socialises individuals to - b - organization more successfully than conventional induction -which is training in "pivotal" role behaviour only. 1.9 Relevant Research As part of a larger study in a Texas Instrument microminiature circuitry department, Gomersall and Myers (1966) interviewed employees and found that anxiety hindered the implementation of a job enlargement programme. An anxiety questionnaire was developed and ad­ ministered to old and new employees in the department. The questionnaire identified an anxiety in new employees that stemmed from the "unpredictable and sometimes threatening new world of work" (pg. 65). The authors then hypothesised that job competence may increase if anxiety was decreased. Anorientation programme for new employees was designed to this end. Employees in the experimental group participated in a one day programme designed to overcome anxiety. The orientation programme emphasised four points: "your opportunity to succeed is very good; disregard hall talk; take the initiative in communication; and get to know your supervisor." Both the control group and the experimental group went through the department '.s conventional two-hour orientation which consisted of a briefing on hours of work, insurance, parking, work rules and employee services. Measures of job performance, absenteeism, tardiness, and total hours required in training were the dependent variables. The experimental group had produced more units per hour, were absent and tardy less, and had less training hours than employees in the control group. The desired behaviour was obtained but the study did not test directly whether anxiety was reduced by the experimental programme or whether competence increased as anxiety decreased. Although the role of anxiety was not directly tested, it was assumed to be a mediating variable between the orientation programme and the results obtained. The orientation programme developed by Gomersall and Myers (1966) contained information that Schein (1968) would catagorise as being "peripheral" from the organizations -~- perspective . 'l'Le detail a nd content of the Gomers all and Nyer 1 s (1966) experimental condition orientat i on pr ogramme provides an exampl e of a socially based induction programme . MASSEY NlVfRS !: llM.R.:t -10 - CnAPTErt II The Theory of \vork Adj ustment and Organization Socialization 2 .1 Introduction The theory of Work adjustment proposed by Lofquist and Dawis (1969) has considerable face validity and appears us eful to this thesis. Adjustment to work appears congruent with the concept of successful org anizational socialization and may help in the identification of suitable dependent measures. This chap~er discusses the suitability of the dependent variables that the theory of Work Adj ustment provides. 2 . 2 Overview of the Theo ry of Work Adjustment The basic assumption of the t.beory is that each individual seeks to achieve and maintain correspondence with his environnent . This correspondence can be described in terms of the individual fulfilling the re~uirements of the work environment and the work environment fulfilling the requirement of the individual . A diagrammatic repres entation of the model can be seen in Figure I. The basic assumption of the theory of work adjustment appears compatible with the theories of motivation presently applied to the work situation. however , a detailed comparison is beyond the scope of this thesis . A further assumption of the theory is that the continuous and dynamic proces s by which the individual seeks to achieve and maintain correspondence with his work environment is called work ad justment. This assumption is congruent with the definition of socialization proposed by Cogswell (1968) . - 11 - Co rrespond ence J --~ I Satisfactoriness ! Promote I i~ bi li ties ~4 Ability Heq uircments I Individual !Job ] I I 1,_ i-ve_e_d_s_,l T -y---R-e....;_i_n_f _o_r c_e_r_ System FIGUdE 1 \vork .<1d justment From Lofquist, L. h ., and Dawis , R. V. ''Ad justment to Wo rk" Century Psychology Series 1 969 (pg . 52 ) I - 12 - "Work represents a ma jor environment to which most individuals must relate". Lofquist and Dawis, 1969). A further assumption of the theory is that the stability of the correspondence between the individual and the work environment is manifested as tenure in the job. From ? igure I it can be seen tenure is in turn influenced by the individual's successful correspondence with the work environment as reflected in satisfaction and satisfactoriness or performance. Proposition I of the theory of Work Adjustment states that "An individual's work adjustment any point of time is indicated by his concurrent levels of satisfactoriness and satisfaction". (Lofquist and Dawis, 1969 pg. 50). The theory is suitable to evaluate a socially based induction programme as it provides possible theoretical measures of the general process of organizational socialization. "Satisfactoriness and s a tisfaction can also be viewed as outcomes in the work adjustment process at various points in time during the individual's period of employment. In this sense, they are measures of work adjustment." (Lofquist and Dawis (1969) page 47). If satisfaction and satisfactoriness are to be used as measures of work adjustment and thus organizational socialization some discussion on the relationship between both measures is warranted. 2.3 The Relationship Between Satisfaction and Performance Lofquist and D av~s (1969) define satisfaction as the correspondence between the needs of the individual and the reinforcer system of the work environment. Satisfaction as a measure is considered independent of satisfactoriness which is defined as the correspondence between the abilities of the individual and the requirements of the work environment. (See Figure I). However, both satisfaction and satisfact­ oriness predict and influence job tenure. Vroom (1964) examined job satisfaction and job behaviour relationship studies which he found supported a negative relationship between the two with low job satisfaction being associated -12- - with high job turnover. He also reviewed twenty correlational studies of the job satifaction and job performance relationship which revealed only a small relationship between the variables. Much of the research in this area has been directed at the possibility of a consistent relationship between the two. Ivancevich (1978) reviewed the literature and stated that "the results highlight the inescapable complexity of the performance to satisfaction causal direction". Lifter (1973) suggested that "while overall job satisfaction may be generally independent of job performance, that portion of job satisfaction which is based on job content is related to job performance". The reviews of Vroom (1964) Ivancevich (1978), and t he research of Lifter (1973) lend support to Lofquist and Dawis's (1969) propositions that satisfaction and satisfactoriness (Performance) are moderator variables of each other only. "Proposition IV; Satisfaction moderates the functional relationship between satisfactoriness (performance) and ability - requirement correspondence. Proposition V Satisfactoriness, moderates the functional relationship between satisfaction and need reinforcer correspondence". Lofquist and Dawis, 1969, pg. 53). Recent research tends to support the propositions of Lofquis t and Dawis (1969). Satisfaction and satisfactoriness as measures of work adjustment can thus be considered theoretically independent. The dependent measures can therefore be hypothesised separtely as indicators of work adjustment as outlined in Proposition I of the theory. 2 .4 Hypothesis An induction programme that contains pivotal, relevant and peripheral role behaviours will effect higher measures of satisfaction, and satisfactoriness and lower labour turnover than a conventional induction programme. -1~- Cr~PTER III METHOD 3.1 Experimental Design In this study the independent variable was whether or not a new employee received an improved induction programme. The dependent variables were measures of job satisfaction, satisfactoriness and tenure. The objective of the improved induction programme was to socialize the new employee into the work environment more effectively than employees who had not received the improved induction. Effectiveness was measured by improved job satisfaction and performance measures in the short term (particularly because of the high turnover of the new employees). A control group was required to ensure that any short term changes in the dependent measures reflected intervention of the improved induction programme rather than trial and error socialization. The design is outlined in Figure II. It was planned that subjects in both experimental and control groups would undergo the organizations present induction procedures and receive a measure of job satisfaction after three days employment. The time period days is arbitrary. No past research offers any guidance on this point. It was considered that after three days employment one could obtain a measure of job satisfaction that could be used to measure the similarity between the experimental and control groups. Subjects in the experimental group received the induction programme immediately after the first measure of job satisfaction. Ten working days after the first measure of job satisfaction the subjects in the experimental and control groups received the second measure of job satisfaction. The ten working day interval between the measures was chosen on a logical basis as the literature offered no guidelines. If the time interval between the pre and post testing~ too long it would fail to test the effect of the induction programme. Trial and error socialization would outweigh the influence of the induction in the long term. If the time interval between the pre -15 - and post testing was too short the test could fail to accurately measure the short term change. The use of a control group raised the problem of possible contamination of that group if the experimental group was run prior to, or concurrently with it. The possible contamination effect of an induction programme with a social psychologicaL content in a social environment like work cannot be overstated. Obtaining measures from the control group prior to the experimental group partly solved the contamination problem. the possibility of the "hawthorne" However, this increased effect as differential treatment between groups could have been observed. COHTROL GROUP Interval EXPERI NENTAL GROUP 'i' i me of Employment 'l'hree ',;Jerking Days After Employment Time One Organization's ~easure of present Job Satisfaction Induction Programme Programme 'l'en Working Days After First Heasure Time Two l·!easure of Job Satis­ faction Measure of Perforanc e Twenty Working Days Between Las t Measure of Control Group and First Heasure of Experimental group Time One Ti me Two Organiza t ion 1 s 1>1easure of Neasure of present Job Satisfaction Job Satis- Induction Additional faction Programme Induction Progra- 1-:easure of Progra!D!l1e Perforance All Job satisfaction measures to be kept in a locked cabinet for the duration of both control and experimental groups to avoid experimenter effect . FIGUHE II Experimental Paradigm - 16- The selection of the employees forming the population from which the control and experimental groups were chosen was of necessity based on the organization1 s criteria for -employment and was thus beyond the control of the experimenter. However, it can be assumed that the same criteria were applied f or both groups and the only variable influencing the assignment to groups was the time of appointment. The variability due to the organizations selection procedures and the history of new employees should have effected both groups equally. The need to obtain measures on the control group prior to the experimental group removed the possibility of randomly assigning new employees to either group. As there was a need to keep the timing of the measurements constant in relation to individual appointment dates earlier subjects were "drip fed 11 into the control group while later subjects were assigned to the experimental group . Although this method of allocating subjects to groups is not normal experimental practice it appeared to be the only practical alternative. Changes in the work environment and the influence of the time of year were not controlled within the experimental or control group due to the "drip feeding" of subjects. In assessing the influence of the induction programme on job satisfaction the design proposed is a .modification of the untreated non-equivalent control group design with pre-test and post-test. (Cook and, Campbell, 1979). The modification is the measurement of the control group prior to the experimental group to avoid contamina­ tion. For the performance or satisfactoriness measure the design is a post-test only design with non equivalent groups. 3.2 Selection of the Organization Interviews with personnel managers of prominent manufacturing firms within the Palmerston North region were held during March/April 1979. Information concerning the nature of the firms• induction programme, the work - 17- climate (interactive versus isolated work), employee turnover, and receptiveness to rese~rch was obtained. 'l'he criteria for the selection of the organization were therefore: 1) the presence of a conventional induction programming; 2) an interactive work environment with a high frequency of communication among employees; 3) the potential for obtaining sixty new employees (size of the organization and the employee turnover within the organization); and 4) the organization's support for the proposed research. Three organizations met the first three criteria. One of the organizations was inundated with research, the other was unreceptive to research in general, but fortunately the third organization gave permission and full support for the proposed research. The target organization had a conventional induction programme that consisted of a guided tour of the factory, information on direct crediting of wages, safety rules, introduction to a training officer and an initial five day training period on the machinery. The induction programme can be classified as being primarily concerned with "pivotal'' role behaviour. The work environment consisted of a large shop floor containing twenty one machine units each operated by two people over the three shifts per day. Machine noise was minimal thus enabling interactive communication between the two operators on each machine as well as frequent communication among other employees. In such a work environment, it was considered that "peripheral" role behaviour was an observable and salient feature of the total role behaviour important to the individual's socialization. A ratio of approximately one supervisor to ten operators was observed. Stringent quality control necessitated supervisor - employee interactive behaviour which was "relevant 11 to the work environment. The organization employed a total of 147 people in the primary production of the product. This total was over three - l .S- shifts per day. The labour turnover for the department was approximately eighty percent per year. The high turnover and planned _full manning of the 10 p.m. to 6 a.m. shift indicated that the desired sample size could be obtained. The organization supported the proposed research and offered to meet material and technical expenses that would be incurred in the development of an induction programme. 3.3 Control Group Subjects The subjects in the control group were new employees in the primary production department of the organization employed after 16 July 1979. Thirty new employees were placed in the control group which ceased on 1 September 1979. The analysable control group was reduced by two due to non-completion of both job satisfaction measures. It was further reduced by ten due to the lack of identification provided on the second measure of job satisfaction. This attrition could be a selection artifact in the control group pre-test pos t -test comparison but the differences between the pre-test scores of those subjects completing post-tests adequately and those not,were insignificant (See Table IV, pg. 30. ) The usable subjects in the control condition consisted of 14 females and 4 males. The age of the subjects ranged from 17 to 48 years with a mean age of 27.4 years. The marital status of the subjects was 5 married and 12 single. 3.4 Experimental Group Subjects The subjects in the experimental group were new employees in the primary production department of the organization employed after 1 October 1979. A total of ten new employees were placed in the experimental group. The allocation of new employees to the experimental group .. ceas.ed on the 1 December 1979. The organization had met its production quota and further employment in the department was stopped. Labour turnover was reduced prior to the Christmas holiday period due to holiday pay. This was not accounted for, with I - 1?- - the subsequent effect of only ten subjects in the experimental group compared to the desired thirty. The .ten subjects in the experimental group consisted of 7 males and 3 females. The age of the subjects ranged from 18 to 48 years with a mean age of 28.5 years. The marital status of the subjects was 3 married and 7 single. 3.5 Setting The research was conducted in a specialized plastic manufacturing company in Palmerston North. The organization had a total of 250 members, was export orientated, and specialized in the requirements of the agrarian sector. As previously stated, the classification of an organization is important for the replication and generalization of research carried out in organizational settings. Pugh et al. (1969) provides a classification of organizations based on three factors; the structuring of activities, the concentration of authority and the line control of the work flow. Table I presents the results of the Analysis of D imens~ons of Organization Structure relevant to the three factor classification. 1'he analysis was completed by the observation of the organization using the analysis of the dimension of organization structure formulated by Pugh et al (1968). The checklist that was used is presented in Appendix r, The data was collected by interviews with relevant executives within the organization during the process of obtaining information for the development of the induction intervention. This method of obtaining the data is the same used in the Pugh et al (1968) study except that direct formal proof was not sought. 'I'he normative data presented in Table I is the original data from the fifty two organizations randomly selected from all organizations in Birmingham employing over 250 people . by Pugh et al (1968). The validity of there norms to the New Zealand situation may be questionable in the classification of the component scales of each factor as the general origin, history and size of New Zealand organizations would be expected to be different. Pugh et al (1971) cites there - ' - .:::.v - contextual variables as greatly effecting the structure of organizations. TABLE 1 Hesults of the Analysis of the Dimensions of the Organization, adapted from Pugh et al (1968) Factor 1 Structuring of Activities Scale Score xNormative Standard J.i ean Deviation Specialization 74 37.77 19.90 Standardization 65 83.88 22.71 Formalisation 12 27.17 11.66 Vertical Span 7 6.71 1.42 Factor 2 Concentration of Autbori ty Organizational Autonomy 15 1b 6.04 Centralization 24 77.48 13 .04 Factor 3 Line Control of i'lork Flow Sunordinate rtatio 10 31.67 23.90 Formalization of nole performance recording 4 7.87 2.64 % workflow superordina tes J2 .4% 5.64% 6.58% Rating Factor 1 Factors 2 and 3 - ~>< High :>X'+lD"""" high Nedium < •tcr > -I a- Low £.X .I Low L.. _, tr Hating High Hedium Low Hedium Hedium Low Low Low High - 21 - The results presented in Table I indicate that there is no clear direction in the component scales used to measure Factor I, the Structuring of Activities. The measurement of the organization by the component scales of factor I indicates that the organization is very highly specialized. It has a medium degree of standardization, vertical span, and a low measure of formalization. The component scales used to measure Factor 2, the Concentration of Authority, indicate a measure of organizational au t onomy equivalent to the normative mean. The very low measure of centralization indicates that the organization has a low concentration of authority. The component scales in Factor III indicate that the organization has a high line control of work the flow. The subordinate ratio is low and the percentage of work flow super-ordinates is high (12.4%). This indicates that the control of the work flow is excerised by line personnel as apposed to impersonal procedures. The component scale formalization of role performance recording, measured low compared to the normative mean. This supports a classification of the high line control of the work flow. Pugh et al (1971) provides seven types of organizational structures that the three factor classification produces. The organization measured using the analysis of the dimension of organization structure formulated by Pugh et al (1968) indicates an organization that is low on the concentration of authority and high on the line control of work flow factors. This classification identifies the organization as being primarily an implicitly structured organization. However, such an organization is generally low on the structuring of activities. The organization measured does have a low measure of formalization which supports this. The high degree of specialization required because of the production technology does not. The organization does not fit conveniently into the taxonomy provided by Pugh et al (1971) but a classification of a specialized, implicitly structured organization would appear adequate for any replication and generalization. - 22 - 3.6 Job Satisfaction Measure The job Descriptive Index (J.D.I.) (Smith, Kendall, and Hulin 1969) was used to measure the new employees job satis­ faction. The check list questionnaire covers five aspects of the work environment, work, pay, promotion, supervision and co-workers. (Sample of J.D.I. in Appendix II). The J. D.I. was selected because it has been used in the New Zealand s€tting previously by Inkson (1977). I t i s a commonly used measure of job satisfaction that contains both affective and descriptive items. (Smith et al, 1974). The measure of affective and descriptive factors is relevant to Lofquist and Dawis's (1969) proposition that satisfaction is a function of the correspondence between the reinforcer s ys t em of the work environment and the individual's needs (pg. 53). The measure of satisfaction must have descriptive items to ascertain the reinforcer system of the work environment and affective items to ascertain how the individual feels about the jobs ability to fulfill his or her needs. Quin and Kahn (1967 7 pg. 456) report that tbe J. D.I. has high convergent and discriminant validity. 'l'he extensive norming of the J.D.I. (Smith et al 1969) also adds weight to the selection of the J.D.I. as a measure of job satisfaction. The J. D.I. has a split half reliability of . 80 to .88 (Smith cited in Fleishman and Bass, 1974). However, no literature has provided a test-retest reliability for the J.D.I. As the J.D.I. was used in a test-retest situation, the inadequacy of this information will be rectified by the comparing of pre and post-test J.D.I. scale scores on the control group. 3.7 Performance Measure Past performance measureshave concentrated on behaviour profiles and supervisory ratings. The organization keeps hourly records for each machine, including the identity of the operators. These records enable an empirical evaluation of satisfactoriness (performance). The daily average of production units accepted to stock (quality control rejects not included) was ascertained for the five days proceeding the J.D.I. post-test. The performance data is a valid measure of the "individual fulfilling the requirements of the work environment." Lofquist . and Dawis (1969 pg. 45). The number of production units that go to stock is the organizations ultimate measure of performance. The average production figure does not account for days that had machine malfunction, resetting, or raw material delays. The seven products being manufactured had different production rates. The standard factory production figures for these products were used to calculate ratios to standardize subjects production figures in order to enable a comparison between treatment groups. The details of the standardization are in Appendix III. 3.8 Development of the Induction Programme The induction programme to be developed was seen as an extension of the organizations present induction. As such, the programme to be developed was orientated to the "peripheral" and 11 relevant 11 role behaviours of the individual to complement the "pivotal'' role behaviour contained in the present induction. This was seen as providing a total induction that would contain all role behaviours and qualify as socialization. Skills, information. and attitudes inherent in "peripheral" role behaviour were investigated first. To this end, the }~edical Research Council Social and Applied Psychology Unit questionnaire (}-:.RCSAPU) (Warr, Cook, and Wall, 197~was administered to a random selection of 19 employees of the organization. An earlier version of the KRCSAPU was adapted by the writer to the New Zealand situation making use of multiple choice responses in a questionnaire format. The }1RCSAPU questionnaire measures the work attitudes of employees. (See Appendix IV) Five scales were chosen for their relevance to an induction programme containing "peripheral" role behaviour: Job Involvement, Job Satisfaction, Anxiety, Trust and Organizational Committment. The mean scores of the five scales were compared to British norms. (Table II) Despite cultural differences and the multiple choice adaptation, the similarites ~~e marked. Comparison Between a Random 0ample of the 1arget Organization (n=l9) and United Kingdom Normative Data on Fi ve Scales of the K~CSA PU Questionnaire l·lBCS.:i PU Scales Heans of Random t•l eans f rom U. K. Sample of Target Normative Data Organization Standard Standard Deviation Deviation Job Involvement 36.23 6.58 35.13 5.46 Job Satisfaction (Total Scale) 71. 85 13. 94 69 .86 14.18 .4nxiety 20.62 6.23 19.03 6.64 Trust 60.08 18 .58 60.48 12.33 Organizational Committment 44.62 12 . 24 44.64 11.45 The Normative Data is fro m an unpublished paper obtained with an tarlier version of the ~RCSAFU questionnaire prior to Warr et al. (1979). From the five scales analysised, the Job Satisfaction and Trust scales were considered to be primarily related to peripheral role behaviour that would include the existing attitudes between present employees of the firm and that this information should be included in a socially based induction programme. It was interpreted from the results of the total Job Sat isfaction scale that existing employees of the firm enjoyed their job. The mean response to item C3, Section C2, 11 .tiow satisfied or dis a tisfied are you with your fellow workers" was "moderatly sat isi i ed 11 • From that it was interpreted tha t existing employees of the firm were happy witt. their work ma t es. From the results of the Trust scale it was interpreted that the present employees of the firm trusted their fellow wo rk mates • . The Trust sub scale, "Ca pability of Peers" Items G9, GlO, Gll was analysed and (x :;; 16.52, Normat ive x = 15.8 ) from that it was interpreted that existing subjects in the sample considered fellow employees to be capable people. The ~ffiCSAPU was used to obtain quas i empirical information about the attitudes that employees hold towards each other. The use of the British norms to interpret the responses of the nineteen employees is ques tionable. however, the information was conside red r e levant to a socially based induction programme. As in the Gomersall Myers (1966) study inf ormat ion was included which would alert new employees to the possibility of 11 jokes 11 being played on by present employees. Gomersall and Myers (1966) also included 11 ini tiative in communication" and "get to know your supervisor 11 skills suggestions in their experimental induction programme. These factors can be considered as aspects of "pe ripheral" role behaviour. In the present study information about these peripheral role behaviours was contained in role play examples. The role plays chosen involved both employee interaction and employee and supervisor interaction each with a negative example followed by a positive example. The role play situations were aimed at alterting new employees to the alerting pos s ibility of the interaction, and the attitudes and skills - c.: u- that would lead to a s a tisfactory outcome. Ar gyris (1965) suggested three categories of individual and interpersonal comp~tence. 'i'hese were owning up, openness, and experimenting all considered relevant to "peripheral" role behaviour in the organization. In this programme owning up was used in the supervisor/employee role play. Opennes s and experimenting were attitudes suggested in information on initiating communication and in the role play of a potential employee/employer conflict situation. Gayne and Rohwer (1969) presented a check list of instructional events which were used to structure the induction programme. Cogswell (1968 ) provided criteria for the structure of setting most conducive to socialization and t he induction programme was developed with these in mind. To ensure uniformity in the present a tion of the induction and practical considerations of time l ed to the developDent of an initial cass e tte r ecording of the induction programme. The r ecording used the writers voice and was fifteen minutes long. 3 . 9 Pilot Study A formal pilot study for the r ecorded induction was considered impractical due to the pos s ible contamination effect within the one available organization. The recorded induction was screened by three senior management employees of the organization and a management consultant, specialized in training. The recording r eceived approval and a number of suggestions were offered • . Video was suggested as being a superior medium and that video would also allow the support of general information concerning the organization. 3.10 Production of Video Induction Tape The video induction tape thus included a pictorial tour of the organization activities by department. The filming was carried out on a Saturday to avoid any effect on present employees. An informative narration was dubbed onto the tape in a 1 . V. studio. Selected employees were filmed in role play situations and an example of a humerous hazing situation was included. It was planned to have a member of the management not often in contact with employees introduce the video tape. However, this proved impractical. The writer therefore introduced the video tape, narrated the pictorial tour of the organization and was filmed in the explanation of the role plays and t ext of the induction programme . All editing and studio filming was conducted in t he Ps ychology Department Video Studio, Mas s ey University. The content of the Induction Programme Video Tape is cont ained in Appendix V • . 3.11 Administration of the J.D~I. The J.D.I. administered to all subjects aft er three days of employment. The training officer directed each new employee to a r oom to complete the J. D.I. Each J .D.I. was prefaced by a direction to provide name and depar tment on the first page of the questionnaire. (See Appendix II) An assurance of the confidentiality was included with only the researche r having access to individual results. The respondents were asked to complete the questionnaire and \ place it in an attached envelope. The complet ed ques tionnaire were kept in a locked cabinet until the end of the ex periment to reduce the possibility of experiment er effect. The second J. D.I. was administered t o all subj ects ten working days after the initial measure. The procedure and questionnaire were identical to the f irst measure. All questionnaires were scored using the scoring presented in Appendix II. 3.12 Administration of the Induction Video Ta pe New employees in the experimental group i ndividually viewed the Induction Video Tape i mmediately after the first measure of J ob Satisfaction. 'l'he Video 'l'ape was administered in the company by the Per sonnel Offi c er . 3.13 Ldministration of the Performance Measures All performance measures were obtained at the · conclusion of the experimental group. The departmental production records were researched for each new employee in the control and experimental groups . The production to stock figures for each employee on the five days proceeding t.be s econd measure of Job Satisfac tion were recorded. The production to stock figures were averaged . Days production tha t included machine down time due to material changes, resetting, and maintenance were not included. 3 . 14 Administration of Job Tenure l,i easures Three months after the conclusion of the experimental group, the personnel files of t erminat ed employees were resea rched to ascertain which control and experimental group employees had left the company and why. 3 . 15 Research Hyoo t hesis Subjects in the experimental group will have highe r post-test scor es on the Job Descriptiv e Index Scales of; ''work" on present job, presnet "pay", opportuniti es for "promotion", "supervis ion" on present job, and "people" on present job, than sub j ects in the control group. Subjects in the experimental group will have higher measures of performance t han subjects in the control group. Subj ects in the experimental group will have longer tenure than subjects in the control group . - 29 - ChAPTER IV RESULTS 4.1 J.D.!. Means and Standard Deviations As the authors of the J DI (Smith et al, 1969) suggest that the five scales should be treated independently, the results have been analysed separately for each of the job satisfaction scales. In addition the dependent variable of the number of units produced has been treated separately. The control and experimental group pre and post-test means and standard deviations for the five job satisfaction scales are presented in Table III. TABLE III Means and Standard Deviations of the Pretest and Post-Test J.D.I. Scales for the Experimental and Control Groups Job Descriptive Control Group n=l8 Experimental Group n=lO Index Scales Mean Std.Deviation Mean Std.Deviation WORK Pretest 29.72 8.24 31.7 13.1 Post test 26.56 9.27 29.7 14.0 PAY Pretest 19.33 5.37 18.7 5.23 Post test 18.28 6.22 19.5 4.84 PROMOTION Pretest 14.22 5.32 12.1 5.63 Post test 16.06 6.00 12.9 7.58 SUPERVISION Pretest 41.44 8.87 40.9 8.71 Post test 38.278 10.5 39.2 10.5 PEOPLE Pretest 35.06 8.05 40.9 8.40 Post test 36.11 10.3 39.1 10.1 4.2 Control Group Attrition As there was a high attrition rate in the control group it was felt necessary to check whether this offered a threat to the internal validity of the study. Two of the subjects - 30- had inadequately completed t he post-test measure whi l e t he r emaining 10 fai led to put names on the pos t-tes t ques tionnaire despite a reques t to do so. Table IV presents the means and standard devi a tions of the pre-test measures of the twelve subj ects excluded from the control group for the abovementioned reasons. TABLE IV Means and St anda rd Deviations of the Pre-test J . D. I. Sca l es for the Exc luded Control Gro up Subjects J. D. I. SCALE Hi:;P.N STD . DEVI ATION \tlORK 31 .1 7 PAY 17.8 5. 25 PRm:OTION 12 . 9 6 . 8 SUPERVISION 46 . 4 4 . 95 PEOPLE 40 .5 5 . 08 There were no significant differences be twe en the pre­ test measur es of the actua l control group and the pre-tes t measures of the twelve excluded control group subjects. (Work , t= .lO,p). 05; Pay , t= . 265 ,p).05; Promo tion, t= 18,p)". 05 ; Supervision, t= 1.32 , p>.05: and People, t= 1.86 , p). 05 ; df= 28) Attrition in the control group i s therefore not a selection artifact and thus appear s unlikely to offer a threat to the internal validity of the design (Cook and Campbell , 1979) . 4 . 3 Visual Data Analysis The mean sco r es presented in Table III were used to construct s catter plots (after Cook and Campbell, 1979) in order to illustrate graphically the differences between the experimental and control groups on the pre and post test scores of the five scales. These are given in Figures III - VIII . The figures indicate that highly significant treatment ef fects are unlikely and tha t int erpre t ation of significance if found could be difficult. y 60 55 50 45 rJ) C::::l ~ 40 0 u :.') E--i 35 (') ~ =- 30 I H rJ') 25 ~ 20 15 10 5 0 5 10 0 ~ FIGUHE III 15 - 31- fr.l 0 0 d> 0 0 0 @ 0 0 ~ 20 25 30 35 40 45 PRE-TEST SCORES Control Experimental gr oup Experiment a l Group ~ Control Group 50 55 60 J. D. I. Scal e - Work pre- tes t against post - test y 60 55 50 (f) 45 ~ 0 u 40 '/) e-; C/) 35 ;..il E--1 I 30 H { /) 0 0.. 25 20 15 10 5 0 5 10 Key 0 ~ FI GUrtl!.: I V 15 -32- ~ 0 20 25 30 35 40 45 50 55 ontro l Gr oup 60 Exper i ment al Group PRE- TEST SO CriES Control Exper i menta l group J . D. I . Scale - Pay pr e- test against post-test 60 55 50 . t2 5 a:; 0 ~40 [-i :f) w 35 H I ~ 30 0 (:... 25 20 15 10 5 0 0 5 10 15 Kev - FIGURE V -33- Experimental Group Control Group co 20 25 30 35 40 45 50 55 60 PHE - TEST SCORBS Control Group Experimental group J.D.I. Scale - Promotion Pre-test against Post-test ;f) !-L.J 60 55 50 ~ 5 D :I) H 40 tl) ;il ~35 H [I) &: 30 25 20 15 10 5 0 -34- \3 experimental 0 cp GJ 0 0 Group 0 0()1)~ tJ ;::,o Control Group 0 (!) 0 . OS ; Pay , t= . S3 ,p} . OS ; Promotion. t= 1 . 12 , p>. OS and People , t= . 73 , p>. OS ) This indicated the need to use covariance ana l ysis in order to test for the significant differences between the adjusted post-tes t means in the experimental and control groups. ~he ANCOVA was was completed using t he statistical package Teddy Bear (1975) on a B6700 Burroughs Computer. Using the ANCOVA programme on the package each J . D.I. scale was tested for skewness, kurtosis and homogeniety of r esidia l variances. The lat ter is obtained by testing for the significance of the correlation between the absolute va lues of the r esiduals and the expected values . If a significant - 37- correlation is obtained it indica tes that the residuals ar e systema tically rela ted to the ex pected value of the post-t est and therefore not homg enious across all points on the expected value. The results are present in Table v. TABLE V The Tests for Skewness, Kurtosis, and Correlation Between the Absolute Values of the Residuals and the Expected Values Corr el a tion of Absolute J . D. I . Scale Skewness P=.05 t:urto s is P=.05 Res i duals with Expected Values \oJORK - 0.76 Si g . - 0.047 NS - 0. 206 NS Pi~Y - 0.044 l~S - 0.003 NS - 0.116 NS PROLOTIO N - 0.558 NS - 0.014 NS 0. 023 NS SUPERVI SIO N - 0.957 Sig. - 0.019 NS 0.062 i~S PEOPLE - 0.092 Si g . 0 .050 NS 0.159 NS No tests were significant on two of the five scales but the work, supervision and p~ople were significantly skewed in a negative direction (p= .05). The significant skewness of the scales is not of primary concern in ANCO VA , but homogeniety of variance and similar slopes in the regression lines are essential. According to Snedecor and Coc~an (1967) one can conceptualise the ANCOVA as a comparison of the regression lines between the pre and post scores in the experimental and control groups as illustrated i n Figures III ~ VII which provide the scatter plots and regression lines for the five J. D.I. scales. These regression lines may differ in slope, in elevation (point of intercept on the Y- or post-test axis) and in t.be homogeniety of their residual variances. ANCO VA tests whether there is a significant difference in the adjusted means between the experimental and control groups which is reflected in the difference between the points at which the two regression lines intercept the Y or post-test axis. However, this test is only meaningful if one can assume homogeniety of the two "' ' -.. .-~c - regress ion lines do not differ s ignifice.ntly. The ;.HCO'Th programme in Teddy Bear (1975) provides a test of the significance of the difference between these two slopes, (based on Snedecor and Cochran 1967). This data is given in Table VI for all five J.D.I. scales. TABLE VI Tests for Significance of Difference in Slope and Variability About the Adjusted Means Between Two Groups Differences in Deviation Difference of 1rror Variance from Hegression Between/ Variance of Levels \Vi th Groups df F p df F p WOltK 1/24 7.13 0.013 17/9 2.33 0.20 PAY 1/24 2.20 0.15 17/9 1.36 0.65 PROl•10TION 1/24 2.65 0.12 17/9 1.47 .56 SUPERVISION 1/24 3.24 0.08 17/9 1.05 0.98 PEOPLE 1/24 0.21 0.65 17/9 1. 25 0.75 The F test compares the between experimental and control group deviations from the regression, with the within group deviations from regression. There is a significant difference in the slopes of the two regression lines in the J.D.I. work scale and the J.D.I. Supervision Scale approached significance. In addition, an F test was performed to test for the significance of the difference in the mean square errors from the adjusted means in the two groups. This data is provided in Table VI. No significant differences were found on e.ny of the five scales indicating similar variability about ·the ad jus ted means in the experime.n.."'ua'l and control group. In light of the above analysis it was decided to use ANCOVA on three of the five scales but it was considered necessary to perform an ordinary analysis of variance on the J.D.I. Work and Supervision Scales. 4.5 Analysis of Covariance Results In performing the t NCOVA for each scale the total sume of squares is broken down as illustrated in Figure VIII and a series of F . tests calculaten . Tests in the Analysis of Covariance Xodel Total Unadjusted Sum of Squares Test 1 l Regression Unadjusted Unadjusted Total Adjusted Total within groups between groups Sum of Squares Sum of Squares Sum of Squares \ ~ Test 2 Test 3 Hi thin Groups Adjusted Adjusted Regression within groups between groups Sum of Squa res FIGURE VIII The adjusted total sum of squares is obtained by subtracting the total regression effect from the unad justed Total SS. 'l'he significance of this regression effect is given by F test 1 in Tables VII - IX for the three scales respectively. The adjusted within groups sum of squares is obtained by subtracting the within groups regression effect from the unadjusted within groups sum of squares . The significance of this regression effect is given by F test 2 in Tables VII - I X. Finally, the adjusted within groups sum of squares is subtracted from the adjusted total and the F test 3 provided in Tables VII - IX answers the question "are the adjusted mean post-test scores (and hence the intercepts) significantly affected by the treatment factors?". No significant differences were found on any of the three scales. While there was a significant overall regression effect (Test 1 in Figure IX) and a significant within groups regression effect (Test 2 in Figure IX) for the J.D . I. Pay , Promotion, and People scales, this did not result in significant differences between the adjusted post-test means. The hypothesis that subjects in the experimental group will have higher post-test scores on the "Pay", ''Promotion", and "Peop-le" scales is rejected. TABLE VII ANCOVA Summary of the J.D.!. Pay Scale Statistic Sum of !D.F. Squares I M.S. F.Test F p Unadjusted Total 877.7 27 Regression Total 404.85 1 404.85 Adjusted Total 427.86 26 . 18.19 1 22.26 0.00007 Unadjusted Within Groups 868.11 26 Regression Within Groups 413.69 1 413.69 Ad jus ted Within Groups 454.42 25 18.18 2 22.76 0.00007 Adjusted Between Groups 18.45 1 18.45 3 1.02 0.32 TABLE VIII ANCOVA Summary of the J. r .I. Promotion Scale Statistic Sum of D. F. H.S. IP.Test F p Squares Unadjusted Total 1193.86 27 Regression Total 503.92 1 503.92 Adjusted Total 689.94 26 26.54 1 18.99 0.0002 Unadjusted Within 26 Groups 1129.84 Regression Within 454.25 Groups 454.24 1 Adjusted Hithin Groups 675.6 25 27 .02~ 2 16.8 0.0003 Adjusted Between 14.34 1 14.34 3 0.53 0.47 Groups - 4 1 - TABLE IX ANCOVA Summary of the J.D.I. People Scale Statistic Sum of D.F. M.S. F.Test F p Squares Unadjusted Total 2794.11 27 Regression Total 2033.92 1 2033.92 Adjusted Total 760.19 26 29.24 1 69.56 Unadjusted Within Groups 2736.68 26 Regression 'd i thin Groups 2040.44 1 2040.44 Adjusted Within Groups 696.24 25 27.85 2 73.27 Adjusted Between Groups 63.95 1 63.95 3 2.23 4.6 Analysis of Variance It appeared from visual scanning of the "work'' and supervision" scatterplots (Figures III and VI) that a differential effect might exist for high and low pre-test scores. Because of this it was decided to divide subjects into low and high scorers on their pre-test scores and a o.ooo o.ooo 0.142 2 x 2 ANOVA was performed (Treatment by Pre-test Scores) on each scale to test for significant differences in the post­ test scores. The A~OVA was completed using the statistical package Teddy Bear (79) on a B6700 Burroughs Computer. For both scales the analysis of variance between Low and High scores was significant ("work"; F (1)= 16.39, p= o.ooos, "Supervision''; F (1)= 7.54, E-=0.01). This was as expected. However, the analysis of variance between experimental and control groups did not reach significance ("work"; F (1)= 0.04, p= 0.84, 11Supervision11 ; F (1)= 0.12, p= 0.73). The hypothesis that subjects in experimental group will have higher post-test scores on the "work" and "Supervision" . .. ~~- scales is r e jected. The interaction between high and Low scores and experimental and control groups was significant for both scales. ("work"; F (1)= 9.8, p=0.004, "Supervision"; F (1)= 8.7, p= 0.007). The means for each treatment by liigh/Low classification for the Post-tests of the J.D.I. "work" and "supervision" scales are presented in Figures X and XI. The pre-test means on the high and low scorers in both the experimental and control groups are included on the graphs. These graphs are included to aid interpretation. In the control group on both the "work" and "supervision•• scale there appears to be a regression to the mean on the post-test, high pre-test scorers scoring lower on the post-test, while low pre-test scorers score higher on the post-test. In both scales with the experimental group, the high scorers remain the same, while the low scorers score slightly lower. The significant interaction obtained may have been due to the changes in the control group coupled with changes in the low scorers in the experimental group. 4 . 7 Production . Table V presents the means and standard deviations of the average daily production in units of stock over five days prior to the post-test J.D.I. for both groups. 40 35 30 25 20 15 10 5 0 high Low ~--------------------~------------~ Control Group Experimental Group = Post-test o = Pre-test t. FIGURE IX The High and Low Means for the Post Test Scores on the J.D.I. work scale 50 45 40 35 30 25 20 15 10 5 0 ~- - Control Group High Low Experimental Group = Post-test 0 ---- = Pre-test ~ FIGURE X The High and Low Heans for the Post Test Scorers on the J.D.I. Supervision Scale , ' - - :!. ""': - TABLE X The Means and Standar d Deviations of Aver age Daily Production Figures for Units to Stock for each Treatment Group Tr eatment He an Standard Deviation Control Group 7486 . 9 743 . 46 n = 11 Experimental Group 7531.1 906 . 25 n = 4 Production figures could not be ontained for nine subjects in the control group and six subjects in the experimental group as the identification on the production sheets was i nadequa te. The subjects ' supervisor had not filled in the machine re co r ds . No significant difference was found between the production figures of either treatment group. (t = - . 097 t .os . The hypothesis that subjects in the experimental group wi l l have higher measurers of production is rejected. 4 . 8 Tenure Table XI shows the number of subjects in the control and experimental groups who cee.s ed employment •.d thin 3 months of tenure , and their stated reasons for doing so . No significant difference was found in the proportion of subjects leaving in each treatment group with three calender months of employment . (x2 (1) = 0 . 754 , p .os . The hypothesis that subj ects in the experimenta l group will have longer tenure is r ejected . - 4.5 - Length of Tenure and Stated Reason for the Subjects in Experimental and Control Groups who Ceased Employment Within Three Calender Honths or 90 Working Days Tre a tment Stayed I, eft VJorking Sta ted i-\e aso n Days for l eaving Experimenta l Group 7 3 1) 28 Obtained other employment (Total n = 10) 2) 39 Re turned to University 3) 28 Re turned to other = 31.7 Control Group 15 3 1) 45 health Reasons (Total n - 18) 2 ) 30 No reason given 3 ) 29 Moving to Auckland = 34.7 4.9 Summary There was no sginificant differences between the J. D.I. pre-test means of the actual control group and the pre-test measures of the twelve excluded control group subjects. No significant differences were found between the pre-test means of the experimental and control J.D.I. scales using a T. test. No signif~cant differeaces were found between the post-test meBns of the control and experimental group also. bn ANCOVA statistical package was used to test for the significant differences between the adjusted post-test means in the experimental and control groups. Each scale was tested for skewness, kurtosis, and correlation between the absolute values and the expected values. The "l:lork", "Supervision" and "People" scales were significantly skewed in a negative direction. The scales were tested for the significance of the difference in slope and variability about the adjusted means between the two groups. There was a significant difference between the slopes of the two job ~ ,.. - '"1.0 - r ee re ss io n lines on t he 11 vJOrk scf!le and tbe "supervi sion11 scale approached significance. An Analysis of Covariance was carried out on the J . D. J . "Pay" "Promotion and "people " scales and no significant difference was found between the groups on the post-test • . !..n Ana lysis of Variance was ca rried out on the J .n.r . "i·lork" "Supervi s ion" scales . No s ignificant difference was found bet ween the experimental and control e roups po s t - t est sco r es on t hese scales . however , a significant intE·r act ion was fo und betv1een high/10\.,r pre - t Gs t s cor e rs po s t-t E.s t scores and experiment a l a nd control groups . No significant difference was found between t he experiment al or control group subjects performance me asur es or length of tenure . ChAPTER V DI~CUSSIOH From the results presented in chapter four, none of hypotheses have been substantiated. It was predicted that the experimental intervention would l ead to an increase in the scores of the experimental group subjects on the Job Descriptive Index scales of "vlork", "pay", "Promotionn, "Supervision" and "People". The socially based induction video t ape did not effect a significant increase in job satisfaction. The significant interaction effect for each treatment by high/low classification on the J . D.I. pre tes t sco res for the "work" and "supervision" scales indicates that experimental subjects with r elatively positive at t itudes towards these two variables did not show the same r eg r ession to the mean as the control group subjects. The induction video tape may have sensitized thos e subjects with negative attitudes to the "work" itself and to "supervision" to possible negat ive aspe cts on the job such that their attitudes toward work and supervision became more negative . The significance of the overall inte raction between treatment and higrJlow classification indicates that training programmes may not have a linear effect and future research may have to take this into account. A design that classifies subj ects into high , medium, and low categories on thei r pret est scores would enable the evaluation of any non-linear effect that may occur. As stated in the r esults chapter, the attrition of subj ects in the experimental group was high because of the non-completion of the questionnaire and the termination of the experimenta l group due to the employment conditions in the organization. 'l'he small number of usable job satisfaction measures decreased the ability to compare a representi tive number of subjects in both experimental . . and control groups. however , the greatest limitation in the compar i son of the experimental and control groups ' job satisfaction scores may well have been due to inadequate timing of the J . D.I. measurement. No literature gave any direction RS t o the optimum timine of the two J. D.I. measurements within the established quasi-experimental design used. The timing of the subjects employment date to detect differences between subjects or initial differences between groups. The J.D.I. post-test may have been too long after the experimental intervention to measure any effect of the socially based induction video as it was hypothesised to influence initial socialization. The J.D. I. was chosen as it is a widely used measure of Job Satifaction and the check-list is easily administered. The J.D.I. has also been used in New Zealand previously (Inkson, 1977). As there is no reported re-test reliability for the J.D.I., this is an uncontrolled factor in its use in a re-test situation. The use of the pre-test to account for initial differences between groups may not have been adequate if the re-test post-test was not reliable. The measurement of the "production to stock 11 for each subject as the organization's criterion for job satisfactoriness or performance appears to be operationally well defined. however, the subjects' data was incomplete, and with data from only nine subjects in the control group and four in the experimental group, the comparison is questionable. Apart from the limitations of the number of subjects in each groups initially, the non-completion of production records by supervisors reduced the usable production records to those analysed. This problem should have been forseen but the practical constraints of the research setting could not be overcome. The comparison of the control and experimental groups' length of tenure showed that the experimental intervention did not lead to a significant difference between the experimental group and control group subjects in the proportion of subjects leaving. According to Lofquist and Dawis (1969), tenure is the ultimate m~asure of work adjustment. Again, the limited.number of subjects allocated to the experimental group may have affected the comparison. From the available subjects, it appears that the experimental intervention did not effect any significant difference in the tenure of the experimental and control group. Tenure is of course influenced by the level of unemployment and as an indicater of work adjustment it is confounded by the prevailing economic climate. The three month tenure period was prior to Christmas for the control gr oup and included the Christmas break for the experimental group. The termination of employment during this period may have been affected by holiday pay and statuary holidays. These factors were also thought to have contributed to the small number of subjects in the experimental group initially. The socially based induction video tape did not result in any changes to the outcome of the dependent variables chosen. In light of this a more detailed discussion of the content of there tape is warranted. Induction has been referred to as a method of training and it can therefore be assessed as a training method. G.ayne and Rohwer (1969) presented a check list of instructional events which serve as criteria against which training programmes can be assessed. These criteria are: 1) Gaining and maintaining attention 2) Directing and preparing the learner 3) Presenting the stimuli 4) Promting and guidance 5) Conditions of responding 6) Feedback 7) Retention 8) Transfer Gaining and maintaining attention was assured by the use of a noval medium and the presence of role plays that would be relevant to the new employee. ''Directing and preparing the learner" to the content of the induction programme was a function of the timing of the viewing of the video three days after employment. "The presentation of the stimuli" in the use of video is both pictorial and verbal. The added advantage of this method of presentation is that the individual is exposed to the stimuli on an individual basis which minimises distraction. "Prompting and guidance" of a new employee was inherent - 50- in the use of a negative role play example followed by a positive role play example of behaviour. No conditions of responding to the medium were provided, and thus no feed­ back was given. This is a definite drawback of the method used. Without continual and direct observation of the new employees while viewing the video, presence of the desired overt responses could not be ascertained for purposes of helpful feedback. Such monitoring was impractical. "Promoting retention" of the stimuli presented was covered by the noval medium. In addition, the realism of the situations and stimuli provided, and the anxiety that may have been produced by the~, was also considered important. The sequencing and timing of the stimuli is an important factor affecting the trans f er. After three days employment the new employees were reasonably familiar with their immediate physical surroundings. This allowed for the presentation of stimuli associated with the "peripheral" role behaviour anticipated in their job .• The sequencing of stimuli in the induction video tape was organized in a perceived chronological order associated with the new employees• training so as to assist transfer of training. While a detailed eva luation of the extent to which the video actually achieved the training objectives outlined by Gayne and Rohwer (1969) was not undertaken, the general analysis and above discussion of the tape point to its having face validity as a training device, although it lacks the interactive element . Gogswell (1968) considered that both the novice and the agent should have clear role definitions, a clear understanding of the purpose of the socialization and that the action of the agent of socialization should not be repetitive in native. Video, as a media, is obviously associated with the disemination of information and the novice, as viewer, defined as the recipient of this information thereby ensuring clear role definitions. 'l'he purpose of the content of the video tape was clearly stated at the start of the tape. The standardization of the video presentation ensured that reptition was kept to a minimum. - 51- While a group structure in t be socialization s e tting is considered desirable, in this study new employees were exposed to the video tape on an individual basis. This was necessary in order to standardize timing for the experimental and control groups. Despite this limitation of the socialization imposed through a design requirement the new employee/induction video relationship appears to have structural properties conducive to socialization. Deviations from Cogswell's (1969) ideal settings for socialization appear reasonable in view of the medium and content of the induction video tape. Other training media may be required to improve the socially based induction programme. Additional m~terial that would enable new employees to respond immediately to the medium and feedback to be given, would be an advantage. A socially based induction programme should also be conduated with a group structure as outlined by Cogswell (1969). however, once one includes response feedback interaction and a group situation for training purposes, the control over the intervention is lost which makes i t difficult to evaluate the intervention using a quasi­ experimental design. There were a number of difficulties in applying good experimental design in an organiza t ional setting. The 'drip feed' selection and allocation of subjects to control and experimental groups was necessary because the organization did not employ large groups at once. As the experimental hypothesis required time controlled measurement and experimental intervention in relationship to the individuals' appointment dates, the design could not include a group setting for the socially based induction programme. Trial and error socialization may have outweighed the effect of the socially based induction programme despite timing the measurements to be constant in relation to ind~vidual appoint­ ment dates. The possibility of the contamination of the control group by the experimental group if it was run prior to or concurrently with it led to a control group first design. -5 ~ - This design increased the possibility of the "hawthorne" effect but it was considered that contamination would be the greater problem. Under the circumstances the modified untreated non-equivalent control group design with pre-test and post-test was the most effective design to control for maturation by trial and error socialization and control group contamination within the organization. However, even with an intervention that has face validity as a training device with structural properties conducive to socialization, and a design that accounted for the problems within the organizational setting, none of the hypotheses presented were accepted. It remains to examine the theoretical and empirical adequacy of the dependent me asures chosen. The propositions provided by Lofquist and Dawis (1969) identified the theoretical dependent variables of "satisfaction" and "satisfc.ctoriness". These factors were operetionalized by Lof quist and Dawis (1969) with the Hinneso ta Satisfactoriness Scales and the 1v1 innesota Satisfaction Questionnaire. In this study the theoretical dependent variables of work adjustment were operationalized using the J. D.I., direct performance measures, and l ength of tenure within a specified time period. The operationalization of work adjustment in this study appears acceptable but it may not have been sophisticated enough to adequately measure the · extent of . the variables postulated by Lofquist and Dawis (1969). Gomersall and Eyer's (1966) suggested anxiety as a dependent variable when evaluating induction programmes even though they did not directly test it. Anxiety may be a moderating variable such that it must be reduced before job satisfaction can improve. The J. D.I. does not have an anxiety scale. The }lliC Social and Applied Psychology Unit Questionnaire developed by Warr et al (1979) has a self rated anxiety scale and using it in addition to measures of job satification may have produced data relevant to the effect of anxiety on other aspects of job satisification. Lo;fquist and Dawis (1969) proposed that "satisfaction is a function of the correspondence between the reinforcer system - 53 - of the work environment and the individual's needs, provided that the individual's abilities correspond with the ability requirements of the work environment." A measure of anxiety could be an additional indicater as to whether or not a correspondence existed between the reinforcer system of the work environment and the individual's needs. The theory of work adjustments was used as a theoretical basis for the above of dependent measures because of its similarity to the concept of organizational socialization. however, future studies should consider using the now v1ell validated HHCc>APU scales dev eloped by ·,J arr et al (1979) to measure work attitudes anxiety and aspects of psychological well being resulting from socially based induction programmes. Future socially based induction programmes need to include additional training media that will enable the novice/new employee to respond to the stimuli so that f eed­ back can be provided. 7he design us ed to evalua te future programmes will require a differentiation of subjects on the basis of pre-test scores to account for any negative sensitization which may occur as a r esult of the programme. In addition the possibility of the socially based induction having a negative effect on subjects in the experimental group could be investigated with regard to subjects' "Activeness" or "Reactiveness". Lofquist and Dawis (1976) proposed that the personality of the employee can be catagorised as "active or "reactive" in relationship to the way they adjusted to the work environment, based on the historical data of the individual. The catagorization of subjects on there dimentions could be a useful way of evaluating the possible non-linear effects of induction as discussed previously. The measureme nt of the "satisfaction'' dependent variable should include a measurement of anxiety as it may be a moderating variable as sug gested by the Gomersall and Myer's (1966 ) study. The collaborative nature of the research and the organization has meant that the experimental design used -54 - and the na ture of the data collection have not been as rigid or as s ystematic as normally defined in research. Warr (1976) outlines the concepts and context of action research.0uring the pr.ocess of this research, The author has become involved with the organization used to test the experimental intervention to a greater extent than this thesis indicates. f uture research in this area could benefit from action res ea rch procedures rather than attempting to place a rigid research method into an organization which results in the pitfalls tha t this the s is exemplifies. -5 :J - P.PPUHJ I X I l~o. 1 2 3 4 5 6 7 8 9 .tmalysis of the Dimensions of the Organization Structure Adopted from Pugh et al. 1968 Specialization Example of Title Public Relations and and Advertising Sales and Service Transport Employment Training Helfare and Security Buying and Stock Control Maintenance Accounts . Activities Develop, legitimize and symbolize the organization's cha r ter x Dispose of distribute and service the output x Carry outputs and resources from place to place x 1\ cquire and allocate human resources x Develop and transfer humcn resources x J,; aintain human resources and promote their identificetion with the organization x Obtain and control materials and equipment l-1aintain and erect buildings and equipment Record and control financial resources X X X 10 Production control Control workflow x 11 Inspection Control quelity of materials and equipment and outputs x 12 }~ ethods Assess and devise ways of prod- ucing output x 13 Design and Development Devise ne\v outputs, equipment and processes x 14 Organization and Develop and carry out administ- methods rative procedures x 15 Legal Deal with legal and insurance requirements x 16 Market research Acquire information on operational field x No. 1 Publicity Staff Public relations Customer relations Display Publicity by product Overseas relations No. 2 Sales or service Pricing and order - 5 6 - Sales by customer or product Sales records Export Sales Service by customer or product No. 3 Drivers Dispatchers X X X X X X X X X X X Ad~inistration and planning x Drivers by vehicle or product x Dispatch specialized by product x Travel and excursions x Planning and administration specialized by product x No. 4 Role specialized for part of organization Role specialized for whole of organization Role specialized by type of employee or process x Administration/records x Interviewers x Role specialized by type of employee and process No. 5 Operative training Apprentice training General education Clerical training Management training Sales training No. 6 Security Staff Nurses Canteen staff Welfare Officer Safety Officer Fire Service Sports and social 0 ther medical Hagazine editor Suggestions Officer No. 7 - 57- X X X X X X Storekeeper x Buyers x Storekeepers specialized by product, materia l , or process Stock control x Buyers specialized by product or material Stock controllers specialized by product, material, or process Administrator Administrator specialized by particular material, etc. No. 8 Engineer Machine maintenance Building maintenance Electrical maintenance - 5o- X X l1achine maintenance specialized by process etc. x Hew works force Surveyors or architect Instrument maintenance Research into maintenance Electrical maintenance specialized by process, etc. No. 9 Wages clerk Costs clerk Ledgers clerk Cashier Financial accounting Costing specialized by product, factory, etc Financial data processing Salaries payment Auditing Budgeting Cost follow up No. 10 Progressing Planning and scheduling Progressing specialized by process etc. Scheduling specialized by process etc. Machine loading X X X X X X : :x - SS- Uo , 11 Product inspection Product inspection specialized by stages Raw material control Laboratory test of product Division of raw material Inspection standards Policy and adminis t ration of inspection No , 12 Vlork study Work study specialized by process Hethods Policy and administration Process planning Production engineering Layout Draftsme n No. 13 New product research Drawing office Process and equipment research New product research by product Di vision into mechanical and electrical Pure research ~dministration of research No. 14 Statistics clerks Organization and methods Subdivision of statistics Filing and post Committees and policies X X X X X X X X X X X X X No . 15 Legal or insurance Share registrar - 0.::>- Legal section subdivision Legal inquiries No . 16 t-;arket research Market research specialized product Economic analysis X X X - Cl- S TA ImA rlD I f. A '1'10 N This appendix gives a list of possible procedures, each rated for their degree of standardization. A high score means highly standardized and a low score less standardized. All multiple-answer questions have been converted into biserial items. Inspection Score Frequency (0-none, 1-haphazard, 2-random sample, 3- 100%) Range (0-none, 1-some, 2-all new, 3-all) Method (0-none , 1-visual, 2- attributes , 3- measurement) Type (0-none, l-one of raw materials , process or final inspection , 2-process final inspection, 3-raw ma t erials process final inspection Speci a l inspection process, e .g. statistical quality control Stock control Stock taking CO-never taken, 1-yearly, 2-semiannually , 3-quarterly , 4-monthly, 5-weekly, 6-daily Operat ional control Firm plans (0-1 day, 1-week , 2-month, 3-quarter 4-year, 5-over one year, 6-per manent Scheduling (0-as needed, 1-monthly, 2-weekly , 3-daily, 4-continuous) Progress checking (0-none, 1-irregular, 2-regular) Haintenance (0-no procedure, 1-breakdown procedure, 2- mixed , 3-planned maintenance, 4-programmed replacements) 3 3 1 2 1 3 4 2 3 Finsncial control Type (1-whole firm, historical, 2-job costing 3-budge ting, 4-standard costs, Score 5-marginal costs) 3 Range (1-whole firm, 2-one product, 3-some products, 4 - all product s 5- all activities) 1 Comparison with budgets (0-none , 1-yearly 2-half-yearly , 3- quart e rly, 4 -monthly, 5-weekly, 6 - cont i nually) Peo ple : control l Definition of operative's task (1-custom, 2-apprenticeship of profession, 3-manuals, 4-rate fixine, 5-time study, 6- wo r k study , 7-\wrk study and task description) 1 Work study (0- none , 1- some direct workers , 2- all direct worker s , 3- a ll direct workers oper a tives , e . g . maintenance etc ., 4-all dir ect workers operatives clerks) 0 Job evaluation Discipline (set offenses) Discipline (set penalties) Discipline (procedure for dismis s ing staff) Salary and wage review Personal reports by superviors Staff establishment Labour budgets Communication 1 l l 1 1 1 Decision seeking (0-as needed, 1-semistandardized 2-standardized, 3-product justification) 1 Decision conveying (0-as needed, !-semi- standardized, 2-standardized 2 - 03 - Ideas Sco r e Re s earch and development (0-none, 1-development as needed, 2-development department, 3-development programme, 4-research and development department, 5-research and dev~lopment programme) 1 Obtaining ideas (0 ,1,2,3 74,5-number of the following that the organization does: conference attending, conference r epor t ing, periodicals circulat i.on, periodicals r eporting, suggestion scheme) 0 Materials Ordering procedures (0-as needed, plans, 2-datum stocks) 1-production Buyer's authority over what to buy (limited) Buyer's authority over whom to buy from (limited) Buyer's authority over how much to buy (limited) Procedure for buying nonstandard ite~s Procedur e fo r notifying head office of purchas es etc . Bidd