Computerized offices in Canada and the United States: Investigating dispositional similarities and differences

JOURNAL OF ORGANIZATIONAL BEHAVIOR, Vol. 9 77-96 (1988)

Computerized offices in Canada and the United States: Investigating dispositional similarities and differences

URS E. GATTIKER AND TODD W. NELLIGAN The University of Lethbridge, Alberta. Canada

Summary A scale measuring computer dispositions was used in order to examine whether computerization of work affects employees similarly or differently in Canada and the United States. The study explores the possibility that employees may differ intraculturally, that is, according to gender, hierarchical level in the firm, and whether they work with intelligent work-stations or main-frame terminals. The data indicates that intracultural differences are apparent according to type of computer used, gender, and hierarchical level of the respondents. What concerned US. female respondents about computer-mediated work did not necessarily apply to their Canadian counterparts.

Introduction Research which looks at intracultures and intercultures simultaneously to investigate peoples’ dispositions toward computer-mediated work is rare. This is the case even though internationalization of business requires researchers to investigate dispositions across countries’ boundaries (Adler, 1983). The investigations that have been done place their primary emphasis on differences as opposed to similarities and differences. Karsh and Cole (1968) postulated that as countries industrialize, attitudes become more similar. Bhagat and McQuaid (1982) have also argued for future research to look at groups of employees doing similar work in several organizations in the countries being compared.

This paper investigates how an individual’s disposition’ toward technology may be similar or different when compared to other office workers in the U.S. and Canada. Specifically, this study looks at a construct of computer attitudes assessing workers’ views about quality of job life’, work effectiveness, control and communication. Furthermore, this study intends to find out if demographic factors may in part explain why individuals’ dispositions toward computer-based technology differ due to the type of equipment used, hierarchical level in the organization and the

Financial support for this research project was provided by The University of Lethbridge Research Fund, under Contract No. 86-1934405, and by a grant from the Alberta Manpower ‘STEP’ Program, Contract No. 11 149, and the Alberta Manpower Post Secondary Training Program. An earlier version of this paper was presented at the 21st International Congress of Applied Psychology, Jerusalem. We would like to acknowledge Cynthia Cunningham for her help in preparing the manuscript. Requests for reprints should be addressed to Urs E. Gattiker, Technology Research Group, School of Management, The University of Lethbridge, Lethbridge, Alberta, T1 K 3M4, Canada.

0894-37961 SS/ 010077-20$10.00 8 1988 by John Wiley & Sons, Ltd.

Received 28 August 1986 Revised 20 March 1987

78 U. E. GATTIKER AND T. W. NELLIGAN

individual’s sex. In other words, the respondents in the two countries may be similar in their dispositions toward computers, however, intracultural differences across groups such as gender may be apparent. Furthermore, even though women may differ from men, women’s concerns may be different in the U.S. as compared to Canada.

Literature Review The approach presented here is interdisciplinary, drawing upon literature in sociology, management and psychology. Studies in three areas have a direct bearing on this research: disposition and work, conceptualizations of computerization and its effects upon one’s workplace, and subjective culture research. In the sections that follow, we provide a brief summary of recent work in the areas just mentioned and show how such work can be applied to the office technology domain.

Toward a dispositional theory of work attitudes and office technology In recent years, almost all research on work attitudes and computerization has been situationally based. For instance, situational variables such as task characteristics, supervision, and ergonomics have been commonly isolated as determinants of job attitudes (Locke, 1976). Rarely, however, are work attitudes formulated as having an endogenous source of variance, one that is reflective of the ongoing state of the person as opposed to just being a product of the situation.

The hypothesis that personal dispositions affect work and attitudinal outcomes can draw support from three separate findings. First, there is substantial individual variation in the perception of tasks and work situations that have identical formal job descriptions (O’Reilly, Parlette and Bloom, 1980). There is enough ambiguity in work situations which allows individuals to interpret them to fit their disposition. Second, there is substantial evidence linking one’s disposition and mental health/stress (Kahn, 1981; Schuler, 1980). A third piece of data supporting the dispositional perspective is the fact that one’s disposition tends to be stable across situations and time (Staw and Ross, 1985; Staw, Bell and Clausen, 1986).

An individual’s disposition about work with a computer Staw and Ross (1985) have argued that although distinctions are sometimes made between the constructs of personal dispositions, traits, personality, and individual characteristics, these terms are used almost interchangeably in the literature. Some ,research has looked at an individual’s disposition and its influence on attitudinal outcomes (e.g. Staw and Ross, 1985). Seldom, however, have researchers made an attempt to assess an individual’s dispositional state, using respondents with several years of work experience with computers (Gutek, 1983).

‘Quality of work life encompasses, but is not necessarily limited to, an individual subjective evaluation of one’s work and career situation. As such, job features affected by computers, career success and stress, as well as commitment are part of the overall construct of a person’s quality of work life.

’In this paper, disposition is a self-perceived construct. It includes, but is not limited to, one’s predispositions about working with computers. In other words, a self-perceived disposition represents the evaluative side of one’s attitudes toward computers and career success. An attitude is generally seen as a disposition to respond in a favorable or unfavorable manner to an object (Oskamp, 1977, pp. 2-12). This paper concentrates on the evaluative aspect of dispositions which has been increasingly stressed in the literature.

TECHNOLOGY MANAGEMENT 79

Working with technology is often defined as the physical activities, combined with the intellectual and knowledge processes by which materials in some form are transformed into output, using the computer as a tool to perform tasks (Roznowski and Hulin, 1985). Rousseau (1979) has noted that these processes are highly interdependent. For example, working with computers in an office may involve converting input such as financial data into outputs such as a balance sheet. Performing tasks with computer-based technology is a complex process which is likely to be influenced by the person’s disposition toward such technology (Staw et al., 1986).

Working with computers Gattiker (1984) stated that an individual’s disposition toward working with computers may be of great importance as to how one perceives attitudinal outcomes. The individual’s cognitive evaluation of computers in the workplace could help explain differences based on dispositions. A positive disposition toward working with computers would be a desirable state since some researchers have argued that the effective use of technology requires employees to feel comfortable with it. For instance, one recent study showed that personal computers were perceived as being most helpful in improving work effectiveness and the quality of job life (Gattiker, Gutek and Berger, 1985).

Other relevant concerns when it comes to an individual’s disposition are the perceptions of the technology’s impact upon one’s performance and the technology’s exercised control over work as perceived by employees (Salzman and Mirvis, 1985). Although computers may be given credit for improving efficiency and effectiveness on the job, they can also provide new avenues of control for management, which could lead to worker hostility and unrest (Mankin, Bikson and Gutek, 1982). Very little empirical research, however, has been done in this area (e.g. Kling and Iacono, 1984).

Still another related factor is the individual’s disposition toward the computer’s effect on communication within and beyond the firm. Some technologies are more useful for this purpose than others, but the computer appears to have become an important new tool capable of improving communications. Yet, research has shown that people communicating via computers evaluated each other less favorably than did people dealing face-to-face (Kiesler, Zubrow, Moses and Geller, 1985). This result is of crucial significance since a major part of office work involves the exchange of large amounts of data and information (Doswell, 1983; Panko, 1984).

If an individual’s disposition toward technology’s effect upon their work is important, the above-mentioned constructs should help researchers and managers alike to understand better the possible impacts of computerization on people’s perceptions of their work situation. Technological developments in the office make it necessary to study their effects upon the workforce to facilitate their productive use (Carter, 1984).

Responses according to computer technology and subjective culture According to Triandis and Vassiliou (1972), individual differences and similarities fall under the ‘individual system’ which forms a part of the subjective culture system. Subjective culture is a group’s characteristic way of perceiving its social environment. As a result, office workers could differ in their dispositions toward computers based on their demographical characteristics such as gender. As well, some research indicates that groups of office workers such as managers have different expectations, values and beliefs when it comes to work compared to their peers (e.g. Gattiker and Larwood, 1986a).

Triandis (1977, p. 144) suggested several sets of variables which should help to classify subjective culture research. A subjective culture study of organizational settings can be classified based on its investigation of the subsistence system (methods of exploitation of the ecology to


survive such as industrial work), cultural system (man-made part of the environment), social system (patterns of interaction such as roles), the interindividual system (e.g. social behaviors), and, the interest of this paper, the individual system (e.g. perceptions, self-perceived dispositions). The dispositional approach to computerization, therefore, concentrates on the individual system, one important classification in subjective culture research.

Computer technology Some researchers have argued that different types of computer-based technology affect people’s work differently (e.g. Salzman and Mirvis, 1985). For instance, employees working with a main- frame computer may perceive greater control when compared to users of personal computers or word processors (cf: Kling and Iacono, 1984). In contrast, some employees may feel computers are de-skilling their work and may fear losing their job within the organization.

Much attention has been focused on intelligent work-stations and their effectiveness in facilitating communication both within and beyond the organization (Demby, 1985). Spreadsheet and word-processing programs allow preparation of highly presentable reports, often including graphics and figures. These and other new capabilities can improve communication through more effective use of the medium at either end (Pava,, 1983). Yet, applied organizational research studying computers and communication is limited, and most existing research has not compared responses from individuals working with main-frame terminals as opposed to intelligent work- stations (Kiesler et al., 1985). In a study using U.S. respondents from Fortune 500 companies, researchers found that participants who worked with intelligent work-stations felt control was lower than did their colleagues working with main-frame terminals. Additionally, quality of job life was perceived to be higher (Gattiker et al., 1985). It would be interesting to test if these results could be repeated when including organizations of different types and sizes.

Gender An extensive review of the literature on women and work reveals that substantial efforts have been put into discovering gender differences based on workplace segregation and family interdependence (Gutek, Larwood and Stromberg, 1986). Nevertheless, computerization’s possible influence in the workplace, and especially on women, has not gained enough attention (Form and McMillen, 1983). ‘Traditional’ research studying women in the workforce has primarily concentrated on comparing women to men in the leadership domain, satisfaction and possible promotion as well as income patterns (e.g. Larwood, Gutek and Stromberg, 1985). Anecdotal accounts have been published from time to time emphasizing that more research in the area of computerization and possible gender effects is needed (e.g. Morgall, 1983). In an attempt to respond to the need of additional information in this area, our approach will be analytical rather than anecdotal.

Various research data shows that women are affected more often by computerization than men due to their occupations and positions, but these studies have not examined specifically if women evaluate the technology itself differently from their male peers (Form and McMillen, 1983; Gutek, 1983). One recent study, however, found that men thought computers would increase their career possibilities, whereas women were not as optimistic. Women, however, felt more positive about benefits of computer technology for white collar employees (Gutek and Bikson, 1985). The data from this study did not indicate that women’s dispositions toward computers differed from men’s.

A study by Gattiker et al. (1985) found that women perceive computers less positively than men. This difference is even apparent when women who worked with personal computers were compared to men. Women were also concerned about the computer’s effect upon their quality of job life. This and other studies indicate that even though objective changes may have occurred


(Form and McMillen, 1983), these are not necessarily perceived as such by employees. Morgall’s (1983) personal account of experiencing computerization in an office setting is a rare exception, describing in qualitative terms differences between her own and female colleagues’ dispositions toward computers and their male peers. Research investigating these issues from an analytical perspective is very desirable.

Hierarchical level While technologies may, in fact, change work content at various levels, attitudes would not necessarily reflect such objective developments (Gattiker, 1984). One explanation for the lack of differences in organizational levels could be the self-selection process. People tend to choose positions which meet their needs and allow them to utilize their skills (e.g. Brousseau, 1983).

Empirical results have been mixed. For instance, Gattiker et al. (1985) found that managers did not differ from support-personnel in how they perceived their computers. Similar results were reported by Gattiker and Coe (1986) with Canadian respondents. But, the two studies mentioned used different types of organizations: whereas the former paper dealt with Fortune 500 companies all across the U.S., the latter research dealt with companies of different size and industry across Western Canada. It would be interesting to see if a subjective culture study, using similar organizations, would show differences between these groups of employees.

Comparing the United States with Canada Computerization is an international phenomenon which affects management across countries. If management research is to remain relevant to executives, a substantially greater proportion of studies need to go beyond the purely domestic perspective (Adler, 1983). Multinational corporations in their attempt to facilitate their workers’ effective use of the computer need the benefits of such research as much, if not more (Evan, 1976). Potential problems could be avoided by a multinational corporation if managers knew the different and similar concerns their workers from different countries have about computers.

Selection of countries for study The countries chosen for this study were the United States and Canada. Specifically, the study concentrated on California, Oregon and Utah in the U.S. and Alberta and British Columbia in Canada for the following reasons. The states/ provinces have a subsistence system which is mostly industrial with substantial agricultural work. The objective culture of the states/provinces indicates that road systems and tools used to work are again similar.

Since both countries are generally assumed to be similar in their work values, it would be of considerable interest to see if this would also apply to attitudes toward work with computers (Griffeth, Hom, DeNisi and Kirchner, 1985). Thus, whereas two countries have the same language and similar work values, differences in their subjective cultures would result in differences in their disposition structures, values and belief systems.

Summary and conclusion A multiplicity of factors influence an individual’s disposition toward computer-based technology. Efforts to create work systems capable of sustaining good job-person matches must deal with developmental issues such as technological innovations (e.g. Brousseau, 1983). For instance, we need to understand if specific group membership such as being a manager or female may influence a given set of dispositions toward computer-based technology. This would represent a move


toward a model-building effort which would greatly enhance future development in this area of organizational research.

Research Issues Several analyses will be conducted to investigate if the type of computer used can result in varying employee attitudes towards the technology and how gender and hierarchical level in the organization could affect those attitudes (e.g. Form and McMillen, 1983; Salzman and Mirvis, 1985). Additionally, a measure to assess computer disposition developed and tested by Gattiker et al. (1985), and expanded and used in Canada by Gattiker and Coe (1986), will be applied to evaluate responses from two countries.

The following hypotheses were established.

Hypothesis I In agreement with the literature previously cited, the computer attitude scales, measuring quality of job life, communication and control, will not differ for the two countries.

When comparing respondents from two countries with somewhat similar value systems, it would seem natural that differences could not be observed. In other words, subjective culture differences between groups such as managers versus support-personnel may exist, but will not be discovered by such a macro approach. Instead, groups may be different within each country studied (Bhagat and McQuaid, 1982).

Based on the above, the scale means obtained for quality of job life, communication and control will also be analyzed comparing the results of both countries along gender, type of computer, and hierarchical level. Previous studies, however, have reported mixed results (e.g. Gattiker and Coe, 1986; Gattiker et al., 1985).

Hypothesis 2 In accordance with the literature cited earlier, communication, control and quality of job life effects will be perceived differently by users of non-intelligent versus intelligent work-stations in both nations. The groups might also differ according to gender. Specifically, the following two predictions were made: H2a. A respondent’s evaluation of computers will differ based on the type most used in his/ her work; H2b. Computer evaluations will differ according to the respondent’s gender.

Design and subjects Canada A stratified sample of 28 employers was asked to participate in a survey of personnel and their ‘computer attitudes’. The employers represented these groups: 10 were firms selected at random from the Globe and Mail annual roster of Canada’s largest organizations; 12 companies were medium-sized firms from Western Canada; the final six comprised three educational institutions and three government agencies. Except for the stipulation of their locale (Western Canada),


organizations were recruited randomly within each classification. Organizational type was not a variable of interest here. The educational and government

institutions were included because they brought potentially different organizational cultures and constructs of effectiveness to the sample, thus allowing more reliable generalizations from the findings (cJ Blalock, 1984, chap. 4).

All employers were asked to select three to six managers (female and male) and an equal number of support-personnel from a variety of departments, and to distribute a questionnaire to these individuals. In order to avoid influencing selection decisions, organizations themselves determined what they considered to be a ‘manager’. Surveys were returned directly to the researchers.

Of the 380 people asked to participate in the study, 340 agreed and 306 responses were ultimately received (90 per cent). Respondents included both sexes (about 66 per cent were female) and approximately 65 per cent were married. About 150 (44 per cent) were managers and 196 (64 per cent) of 306 participants were computer users, while the remaining 110 indicated other primary technologies such as the telephone or typewriter. The analyses performed for this study were based on the sample of 196 computer users. One hundred and twelve (57 per cent) of these were using a main-frame terminal.

United States Again, a stratified sample of 15 employers was asked to participate. The employers represented these groups: five were firms selected at random from the Fortune 500; five were medium-sized firms from the Western US.; the final five comprised three educational institutions and two government agencies. Except for the stipulation of their locale (Western United States), organizations were recruited randomly within each classification.

The same procedure for respondent selection was used as described earlier. Of the 200 people asked to participate in the study, 185 agreed and 157 responses were ultimately received. Respondents included both sexes (about 47 per cent were female), 61 (39 per cent) were managers and approximately 68 per cent were married. Ninety-five (61 per cent) of 157 participants were computer users, while the remaining 62 indicated other primary technologies such as the telephone or typewriter. The following analyses are based on the 95 computer users of which about 52 (55 per cent) used a main-frame terminal in their work.

Instrument The questions used in this study were originally developed by Gattiker et al. (1985) and expanded by Gattiker and Coe (1986). These authors designed the questions to reflect concerns expressed in the literature about the impact of office technology on workers (e.g. Gattiker, 1984; Kahn, 1981). Their focus is limited since other important considerations such as effect of work skills also have an impact (cf: Salzman and Mirvis, 1985). It is, however, very difficult to measure people’s dispositions toward possible skill changes (cf: Cammann, 1981). Some researchers claim that although skill changes may be apparent to the expert, people often seem to be oblivious to them (Spenner, 1983; Znanieck Lopata, Fordham Norr, Barnewolt and Miller, 1985). The items included in the questionnaire are, therefore, limited in scope and focus specifically on the dispositional approach to computerization.

Respondents completed an anonymous questionnaire to assess their attitudes toward computer-based technology, specifically, how it supports individuals at work, whether they like using their computers, and if such use makes them more effective. A five-point scale, ranging from (1) ‘agree completely’ to ( 5 ) ‘disagree completely’, was provided. The original scale with 18 items

84 U.E. GATTIKER A N D T. W. NELLlGAN

had been developed by Gattiker et al. (1985) using a U.S. sample. An expanded version with 27 items was developed by Gattiker and Coe (1986) using a Canadian sample. The latter instrument was used in this study. Additional questions about one’s type of technology and the percentage of time spent using it were also included. The final section of the questionnaire concerned demographics, asking about annual income, educational background, job title and the like.

Analyses First, confirmatory factor analysis was done with the variables measuring computer attitudes for both countries. This approach has been suggested to see if the factor structures, as obtained in earlier work with different samples, would in fact emerge (Sekaran, 1986). To decide the number of factors for orthogonal varimax rotation and interpretation, eigenvalues (> 1 .O) were considered (Kaiser, 1974). Item scale loadings greater than 0.30 were statistically significant 0, < 0.001), according to the Burt-Banks criterion (Child, 1970). This conservative approach was used to avoid reporting results based on sample characteristics which could not be replicated in the future (cf. Webb, Campbell, Schwartz, Sechrest and Grove, 1981, chap. 3). The statistically significant items were then checked for their item-item and item-total correlation within each factor. Only items which correlated positively with other items in the same factor were retained for the scales discussed below (Nunnally, 1978, chap. 3 and 6). Five scales were constructed by averaging scores from the items which qualified using the above rules.

Second, the scales were then used for t-tests comparing the scale means obtained in the two countries. Additionally, t-tests of the scale means were performed according to sex, type of computer used and hierarchical level (manager versus support-personnel). Additionally, to test if significance, as obtained with the t-test, would reflect a sizeable association in the data, the strength of statistical association was estimated using w2 (Omega) (Hays, 1981, pp. 293-298).

Third, the variables sex and hierarchical level in the organization were used to determine if they would help to distinguish among respondents’ computer attitudes. The type of computer used (intelligent work-station versus main-frame terminal) was also included in these analyses. Univariate and multivariate analyses of variance were done to test for possible differences.

Factors in computer attitudes To obtain the independent factors, orthogonal varimax rotations and reliability analyses were done with the 27 items measuring computer attitudes. Loadings greater than 0.30 were statistically significant 0, < 0.001, according to the Burt-Banks criterion). All 27 items measuring computer attitudes loaded highly enough and were retained to define the following four factors: (1) quality of job life, (2) work effectiveness, (3) communication, and (4) control. Since the factor analyses done here are confirmatory in their character, only the variance accounted for is reported (see Table I). Factor loadings can be obtained from either author.

Except for control in the Canadian sample, the reliability coefficients for computer attitudes are well above 0.70 which has been suggested as a desirable minimum for constructs in the early stages of formulation (Nunnally, 1978, p. 245) (cf. Tables 1 and 2). Therefore, the extended version of the computer attitude measures developed by Gattiker and Coe (1986) can be applied in both countries, achieving desirable levels of reliability.


U S . 1 Canadian differences in dispositions toward computer technology HI stated that the computer attitude scales measuring quality of job life, communication and control will not differ for the two countries. In order to test this hypothesis, the scales were used for t-tests comparing the scale means obtained in the two countries. All computer users were placed into two separate groups according to country. None of the test results were statistically significant. Based on these results, H1 was confirmed (cf: Table 1).

Additionally, an overall scale which contained all attitude scales and the scale labelled work effectiveness were also tested for differences between the means. Neither of them resulted in a significant difference. Tests were also run between the two countries’ female and male, intelligent work-station and main-frame terminal users, managers and support-personnel. Of the 30 tests, only one was significant, namely, the difference between female respondents of both countries and their scale means measuring work effectiveness. The female respondents from the U.S. felt that computers helped to improve their work effectiveness more than their Canadian peers. However, estimating the strength of statistical association as suggested by Hays (1981, pp. 293-296) using o2 revealed a result of 0.002. The association of the data due to the different countries’ effect is, therefore, trivial. These results support Karsh and Cole’s (1968) proposition that highly industrialized countries tend to have similar attitudes. This means that the above results demonstrate that employees in computerized offices in the Western U.S. and Western Canada have similar dispositions toward computer-mediated work.

Computer technology in the office Two types of computer-based technology were examined in these analyses, main-frame terminals and intelligent work-stations (personal computer and word processor). Survey respondents were grouped according to the type of computer used most often. Individual evaluations of computer- based technology were also examined according to gender and hierarchical level (managers versus support-personnel). Multivariate analyses of variance and univariate analyses of variance were used to compare the scores of each of the groups on the four factors.

Hypothesis 2a This hypothesis stated that respondents would differ in their evaluation of a technology depending upon their use of a main-frame terminal or an intelligent work-station. The results of these two analyses for each type of computer are shown in Table 2.

The multivariate test ( F tests of Pillai’s Vfrom SPSSX MANOVA) showed reliable ( F = 2.62, p<0.05) differences between the technology groups on the pattern of their scores on the four factors for the Canadian sample. The univariate analysis of results also reveals a different pattern of responses for how intelligent work-station versus main-frame terminal users evaluated communication. This could mean that the factor communication distinguishes among individuals as to the type of work-station used. The means derived from the scales indicate that individuals working with intelligent work-stations feel communication improves significantly when compared to main-frame terminal users ( p < O . O l , by a two-tail 1-test between the scale means of the two groups).

The results obtained for the U.S. group of respondents, however, were statistically insignificant for the multivariate as well as the univariate analyses of variance done. These results support hypothesis 2a only for the Canadian sample; they do not hold true for the U.S. respondents in this study.

m

o\

Tabl

e 1.

Ite

ms

used

to

defin

e fou

r fac

tors

: Com

Dut

er a

ttitu

des

Fact

or

Item

s 1 Q

ualit

y of

jo

b lif

e

This

equ

ipm

ent i

s fun

to u

se

Usi

ng th

is p

iece

of

equi

pmen

t mak

es m

y w

ork

mor

e in

tere

stin

g I e

njoy

usi

ng th

is p

iece

of e

quip

men

t Th

is p

iece

of

equi

pmen

t ena

bles

me

to d

o in

tere

stin

g ta

sks

at w

ork

The

use

of t

his p

iece

of

equi

pmen

t mak

es m

y w

ork

mor

e en

joya

ble

I lik

e do

ing

my

wor

k w

ith th

e he

lp o

f th

is e

quip

men

t M

y w

ork

wou

ldn'

t be

as

muc

h fu

n if

I cou

ld n

ot u

se th

e eq

uipm

ent

Var

ianc

e exp

lain

ed p

er fa

ctor

Sc

ale m

eans

2 A

t my

wor

k, I

dep

end

a gr

eat d

eal o

n th

is e

quip

men

t W

ork

This

pie

ce o

f equ

ipm

ent e

nabl

es m

e to

do

my

job

mor

e ef

fect

ivel

y ef

fect

iven

ess

This

pie

ce o

f equ

ipm

ent m

akes

my

wor

k ea

sier

Th

is p

iece

of e

quip

men

t sup

ports

me

in m

y w

ork

I am

mor

e ef

fect

ive i

n w

ork

with

this

equ

ipm

ent t

han

I wou

ld

Usi

ng th

is e

quip

men

t mak

es m

e m

ore

prod

uctiv

e Th

is p

iece

of

equi

pmen

t ena

bles

me

to d

o m

y w

ork

fast

er

This

pie

ce o

f eq

uipm

ent e

nabl

es m

e to

do

my

job

mor

e th

orou

ghly

be w

ithou

t it

Var

ianc

e exp

lain

ed p

er f

acto

r Sc

ale

mea

ns

3 Com

mun

i- th

e or

gani

zatio

n Th

is p

iece

of e

quip

men

t fac

ilita

tes c

omm

unic

atio

n am

ong

peop

le in

This

pie

ce o

f eq

uipm

ent f

acili

tate

s com

mun

icat

ion

with

peo

ple

This

pie

ce o

f equ

ipm

ent a

llow

s me

to tr

ansm

it in

form

atio

n to

I pre

fer a

face

-to-f

ace

mee

ting

over

usi

ng th

is e

quip

men

t for

The

use

of t

his

equi

pmen

t has

impr

oved

com

mun

icat

ion

beyo

nd

catio

n ou

tsid

e of

the

org

aniz

atio

n

som

ebod

y el

se

impo

rtant

mat

ters

the

orga

niza

tion

(com

pare

d to

pre

viou

s met

hods

)

Item

-tota

l C

r onb

ach'

s co

rrel

atio

n al

pha

U.S

.A.

Can

ada

U.S

.A.

Can

ada

0.75

0.

71

0.78

0.

76

0.80

0.

81

0.75

0.

55

0.8 1

0.

82

0.78

0.

78

0.62

0.

61

0.92

0.

90

1.78

1.

80

26.1

25

.1

0.36

0.

38

0.55

0.

68

0.58

0.

63

0.49

0.

53

0.45

0.

65

0.63

0.

72

0.52

0.

70

0.52

0.

70

0.80

0.

87

1.48

1.

58

14.1

12

.4

0.59

0.

56

0.61

0.

55

0.5 1

0.

52

0.29

0.

15

0.5 1

0.

58

0.74

0.

71

1-te

st e

com

parin

g pl

coun

tries

R Q

sc

ale

mea

ns

of t

he tw

o 5 m

w * z tl j

5 3 5 n.

s. % r

n.s.

Var

ianc

e ex

plai

ned

per

fact

or

Scal

e m

eans

6.

8 7.

9 2.

43

2.26

4 I f

eel t

his

piec

e of

equ

ipm

ent c

ontro

ls m

y be

havi

our

at w

ork

0.41

0.

40

Con

trol

T

his

piec

e of

equ

ipm

ent

mak

es m

y w

ork

mor

e de

man

ding

0.

46

0.29

M

y pr

oduc

tivity

is c

ontro

lled

by t

his

equi

pmen

t 0.

53

0.41

U

sing

this

equ

ipm

ent l

imits

my

abili

ty to

mov

e ar

ound

0.

35

0.3 I

G

ener

ally

, I p

refe

r to

com

mun

icat

e w

ith e

quip

men

t ra

ther

than

fa

ce-to

-fac

e 0.

36

0.31

If

this

equ

ipm

ent i

s ou

t of

orde

r, I c

anno

t do

my

wor

k 0.

44

0.28

It

is im

port

ant t

o m

e th

at th

is e

quip

men

t be

in u

se th

roug

hout

th

e da

y 0.

44

0.30

0.

71

Var

ianc

e ex

plai

ned

per f

acto

r Sc

ale m

eans

5.

7 6.

8 2.

88

2.89

n.s.

0.61

ns.

Tota

l var

ianc

e ex

plai

ned

52.7

52

.2

Ave

rage

scal

e m

eans

for a

ll at

titud

e fa

ctor

s co

mbi

ned

2.14

2.

11

n.s.

Tabl

e 2.

Mul

tivar

iate

ana

lysi

s of

varia

nce

for a

ttitu

des

tow

ard

com

pute

rs

C m

Mul

tivar

iate

test

s U

niva

riate

of

Pill

ai's

V F-

test

s

Qua

lity

of

Wor

k So

urce

df F

(d

o jo

b lif

e ef

fect

iven

ess

Com

mun

icat

ion

Con

trol

C

* (T

ype

of c

ompu

ter)

C

anad

a:

1 3.

12"

(4,1

72)

4.60

IP

0.55

4.

40IP

4.

26"

U.S

.A.:

1 0.

11

(437

8)

0.15

0.

00

0.17

0.

0 I

C w

ith e

ffec

ts o

f M

rem

oved

t C

anad

a:

1 2.

51"

3.4 I

0.

19

3.73

IP

3.61

" U

.S.A

.: 1

0.33

0.

28

0.00

0.

01

0.24

C w

ith e

ffec

ts of

S re

mov

ed$

Can

ada:

1

2.95

" 3.

29

0.36

6.

87Ib

1.

20

U.S

.A.:

1

0.16

0.

30

0.0 1

0.

21

0.01

Can

ada:

1

2.62

'' 2.

85

0.20

6.

15"

I .23

U

.S.A

.: 1

0.38

0.

39

0.0 1

0.

02

0.23

Can

ada:

1

1.31

(4

,172

) 2.

36

I .98

0.

92

0.89

C w

ith e

ffec

ts o

f M

and

S re

mov

ed8

M*

(Hie

rarc

hica

l Lev

el)

U.S

.A.:

1 3.

90ib

(4

978)

0.

25

0.02

1.

60

5.42

"

M w

ith e

ffec

ts o

f C

and

S re

mov

ed

Can

ada:

1

1.00

0.94

1.56

1.18

0.03

U.S

.A.:

1 1.00

0.16

0.06

1.63

5.62"

Can

ada:

1 3.371b

(4,172)

1.58

0.27

0.94

8.411b

S* (S

ex)

U.S

.A.:

1 2.44"

(4,W

7.601b

1.72

0.78

0.02

U.S

.A.:

1 2.30"

7.52Ib

1.77

1.02

0.00

S w

ith e

ffec

ts o

f C

and

M r

emov

ed

Can

ada:

1

3.51''

0.

05

0.0 1

4-33''

5.14"

Mul

tivar

iate

test

s co

mpa

re th

e tw

o gr

oups

(e.g

. man

ager

s ve

rsus

sup

port-

pers

onne

l) on

all f

ourf

acto

rs si

mul

tane

ousl

y, u

sing

Pill

ai's

V as

cal

cula

ted

by S

PSSX

M

AN

OV

A, w

here

as u

niva

riate

test

s co

mpa

re th

e gr

oups

on

onef

acto

r at a

rime o

nly.

* C

type

of c

ompu

ter u

sed

(mai

n-fr

ame

term

inal

ver

sus i

ntel

ligen

t wor

k-st

atio

n); M

hi

erar

chic

al le

vel o

f re

spon

dent

(man

ager

ver

sus

supp

ort-p

erso

nnel

); S

=

t Loo

king

at c

ompu

ter e

ffec

ts h

avin

g re

mov

ed p

ossi

ble e

ffec

ts b

y th

e pe

rson

's hi

erar

chic

al le

vel i

n th

e or

gani

zatio

n.

3 Loo

king

at c

ompu

ter e

ffec

ts h

avin

g re

mov

ed p

ossi

ble

effe

cts b

y th

e pe

rson

's ge

nder

. 0 L

ooki

ng a

t com

pute

r eff

ects

hav

ing

rem

oved

bot

h po

ssib

le ef

fect

s of a

per

son'

s ge

nder

and

hie

rarc

hica

l lev

el o

n hi

s/ h

er re

spon

ses.

" p <

0.06

. "

p <

0.05

.

gend

er o

f res

pond

ent (

fem

ale

vers

us m

ale)

.

Ibp <

0.01.


Gender of respondent Hypothesis 2b suggested that individual evaluations of computer-based technology would differ according to gender. As Table 2 demonstrates (c$ last row - ‘S with effects of C and M removed’), the multivariate test result for Canada (F test of Pillai’s V = 3.51) shows a highly reliable (’p < 0.01) difference between men and women in the pattern of their scores on the four factors. The univariate tests using scores on the individual factors revealed group differences in both analyses for communication and control (means will be reported in the next section below). There were no reliable differences, however, between a person’s gender and his/ her perceived quality of job life and work effectiveness.

For the U.S. respondents, the multivariate test result shows a difference (p < 0.06) between the men and women as well. It is rather small though, and, as the univariate analysis of variance results revealed, only group differences for quality of job life are recorded. These results confirm hypothesis 2b. Differences in the subjective culture (disposition toward computer-mediated work) between females and males in the two countries do exist: the differences occur in communication and control for Canada; the differences occur in quality of job life for the U.S.

Type of computer Since others had reported differences between respondents’ gender and type of computer used in their work, this study also performed some tests to reveal if such differences would exist in both countries. For Canadians, a person’s sex and the type of computer he/she uses were analyzed simultaneously with a multivariate test of Pillai’s V ( F = 2.51). It showed that respondents differ reliably (p < 0.05) in how they evaluate their computers (c$ Table 2). The univariate tests indicated that the respondents differ in their evaluation of the factors communication and control.

The means derived from the scales indicate. that women perceived increased control when working with a main-frame terminal (p < 0.01, by a two-tail t-test between the scale means of the two groups). Nevertheless, female respondents did not differ statistically significantly from male respondents in their perception of control when judging intelligent work-stations.

These means also show that female respondents differ in their evaluation of computer-aided communication. Women perceived less improvement in communication possibilities from intelligent work-stations than men did (’p < 0.01, by a two-tail t-test of the two scale means). Female respondents did not differ significantly from their male peers, however, when assessing communication with main-frame terminals.

Looking at the U.S. respondents, the multivariate test showed no reliable differences between the gender groups on the pattern of their scores on the four factors, nor did the univariate F-tests in this study. Thus, the intracultural investigation indicates that dispositional differences are washed out for the U.S. respondents, if other effects such as type of computer used are not taken into consideration. This is in contrast to the Canadian sample.

Qpe of computer and hierarchical level Table 2 illustrates that for Canadian survey participants, one’s position in an organizational hierarchy does not really help to differentiate between respondents. In other words, Canadians participating in this survey did not differ along hierarchical lines (manager versus support- personnel) in their evalution of computer-based technology.

Further, Table 2 shows that for U.S. respondents, hierarchical differences were only apparent if effects such as type of computer used and gender were considered (3.90, p < 0.0 1). Looking at the univariate results and testing for hierarchical level only, U.S. respondents felt that control as


exercised by the computer was higher for support-personnel than for managers (p < 0.01, by a two-tail I-test between the scale means of the two groups).

Discussion Computer attitudes in two countries The scales used across the Western US. and Western Canada for this study are justified because, as with earlier studies using a shortened version of the same scale (Gattiker et al., 1985), the factor analysis performed with both countries’ current data sets reveal that the same factor structure was obtained with very similar reliability levels (Sekaran, 1986).

The data also showed, however, that the two countries’ respondents did not differ in how they evaluated quality of job life, communication or control as influenced by computer-mediated work. Except for work effectiveness, where Canadian and U.S. females differed in their assessment of computer effects, none of the additional statistics revealed significant differences between the two countries. One explanation for this pattern could be that the two countries surveyed are somewhat similar, use the same language and share a long land border which facilitates extensive border traffic (cf: Sekaran, 1986). Another explanation, as some research has indicated, is that work values and attitudes are similar; therefore, such results should be expected (Hofstede, 1984, chap. 9). Giving even stronger support to these results is that, at the time of the survey, more than 80 per cent of the users in both samples had worked with the technology for more than 3.5 years; differences due to computer novelty were most likely eliminated.

The above results confirm Karsh and Cole’s (1968) proposition that as nations industrialize, attitudes become more similar. This is not to say that the U.S. and Canada do not differ in their individual cultural systems. Nonetheless, comparing respondents working in similar organizations across states/ provinces which have important similarities with regard to work values and attitudes indicates that overall respondents in this study had similar dispositions towards computers across both countries.

Intracultural differences across groups Triandis (1977) proposed that even though countries may have similar cultures, particular groups within a country may differ in how they perceive their work environment. The following sections will discuss this proposal as it relates to this study’s results.

Computer technology in the office and gender The results reported in this study show that the respondents in the two countries differed in how they assessed computer-based technology based on gender. The Canadian respondents differed in how they assessed the effect of computers on communication and control. Women did not feel communication was improved by the computer, whereas their male counterparts did. Women also did not feel they had as much control over their work because of the computer, whereas men did. These results, then, confirm the ones obtained by Gattiker and Coe (1986).

When looking at the U.S. respondents, quality of job life was the most important factor to distinguish between female and male respondents. Again, females felt that quality of job life was less improved by working with a computer when compared to their male counterparts. This result confirms an earlier study done by Gattiker et al. (1985).

Our data suggest that for men, computers seem to have a more positive effect than for women.


Differences between the sexes are even more significant when considering the fact that respondents from both countries and sexes held similar jobs and worked in similar companies. Gutek and Bikson (1985) claim that opportunity structure and status differences in organizations may be more important in accounting for differential experiences than the technology itself. This claim, however, does not seem to be true when comparing women and men in similar positions across countries. Women’s experiences are still different as this study indicates.

Type of computer used and gender A look at the type of computer used and the respondents’ gender does not lead to significant differences in how U.S. respondents assess quality of job life. This finding does not support the work by Gattiker er al. (1985). One explanation could be that these authors used large organizations only, whereas this study used large and small, public and private firms.

For Canadian respondents, however, once again control and communication dispositions differ, i.e. women feel less in control than men while working with personal computers as well as main-frame terminals. One explanation could be that status differences and opportunity structures may still provide advantages for men as compared to women (cf: Gutek and Bikson, 1985). Even though the study tried to include respondents of both sexes and end-users doing similar work, there still may be differences underlying jobs between sexes, which this study could not account for.

The results in this study indicate that a group’s characteristic way of perceiving their social environment may be eliminated if one looks at gender and type of computer used simultaneously. This is particularly true for the U.S. respondents. In other words, although these respondents differed in how they assess quality of job life along gender, this difference was eliminated if one also looked at the type of technology employed. Looking at the Canadian data indicates that differences in respondents’ dispositions about communication and control remain significant even when looking at type of computer worked with. This comparison shows that the two countries have different intracultural characteristics when it comes to people’s dispositions toward computer-mediated work.

Summary and conclusion Even though the respondents did not differ across countries when comparing the scale means and doing t-tests on them (e.g. Canadian women’s scores on quality of worklife against U.S. counterparts), differences were reported by the univariate tests. None of the differences in computer attitudes, however, were on the same scale for both countries. This would seem to indicate that underlying differences in computer attitudes between the two countries would have to exist. The study’s data does not indicate what may account for such differences.

The results obtained showed that the two countries’ office workers seem to have different computer attitudes. Although comparing the mean scores did not lead to statistically significant differences, looking at gender indicated that in Canada and the U.S. females seem to perceive computers less positively than men. But, whereas the U.S. respondents differed from men in how they assessed quality of job life, Canadian females differed in how they assessed communication and control of computers. The results of earlier studies done in the same countries are confirmed by this study (e.g. Gattiker and Coe, 1986; Gattiker et al., 1985). As a result, there is strong support indicating that Canadian female office workers perceive computers differently than their colleagues in the U.S. These differences seem to fall into a pattern, thereby allowing to identify genuine differences between the two countries, and how certain groups perceive their computer- mediated work (see also Bhagat and McQuaid, 1982).


One explanation for the U.S. result could be that the quality of the job life issue and women and work have been more forcefully and publicly discussed since the sixties, whereas in Canada, forceful and public discussion is more recent (e.g. Gutek et al., 1986). As a result, U.S. women may have a stronger, collective voice about quality of job life issues, thereby leading them to more critical comparison between their own situation and that of their male peers.

Implications for Management and Future Research This study provides strong evidence that research on computer-based office technology should be placed within a larger framework. Any understanding of the effective use of such technology is substantially undermined if employee dispositions are ignored. Future research should continue to explore this issue.

Theoretical implications In evaluating the theoretical implications of this study, one should keep in mind that there are several potential levels of dispositional and intracultural explanations. The mo'st fundamental, yet primitive, is the knowledge about attitudes having some sort of consistency over time and context (Staw and Ross, 1985; Staw et al., 1986). At a second level, the general finding that dispositions toward computers seem to differ according to group characteristics such as gender and type of computer used have some important implications. Specifically, the differences between the two countries seem to fall into a pattern.

Bhagat and McQuaid (1982) argued that genuine differences between countries can be identified if patterns of differences exist among dispositions across countries. Although overall respondents in the U.S. and Canada did not differ from each other substantially, intracultural differences were recorded. Perhaps most important is that Canadian female respondents were less positive about communication and control via computer than their male colleagues, whereas U.S. female respondents were more concerned about the quality of job life than their male colleagues. Looking at computer dispositions in office settings, this indicates that the subjective culture for females working with computers in the U.S., is different from the subjective culture for females in Canada.

Another important level of explanation is to use an etic approach to study differences between countries which is an attempt to describe social phenomena in relatively culture-free, universal terms (Bhagat and McQuaid, 1982). Looking at the study's results from this angle would make one believe that recorded differences in the subjective culture between the U.S. and Canadian office workers indicate different dispositional structures about computerization and work. The question to be answered still is how and to what magnitude these variations may be due to differences of women's roles in the workforce in the two countries (e.g. Larwood et al., 1985). The argument can be made that those results may have derived from differences in subjective culture of the females between the U.S. and Canada.

The argument that U.S. and Canadian respondents did not differ in their assessment of disposition of computer-mediated work, looking at the total sample, suggests some necessary revision of our models of researching different countries. Although research clustering countries may be of value to multinational corporations in terms of policy-making and integrating of the different cultural groups, it may be of minimum value in our attempts to understand the role of subjective culture in organizational contexts (e.g. Griffeth et al., 1985). Furthermore, research

94 U.E. GATTIKER A N D T. W. NELLIGAN

which compares countries with quite different objective cultures and a variety of subjective cultures within each country does not advance our knowledge about cultural differences sufficiently (Sekaran, 1986). Such areas may just represent targets of opportunity which too often yield ‘so what’ results (Bhagat and McQuaid, 1977). Weick (1979, p. 52) has already argued that what is critical in investigating differences and possible interactions in organizational settings is an ability to ‘think in circles’, which means that it would be more advantageous to compare countries with similar objective cultures. Moreover, as in this study, analyzing intracultural groups’ characteristic ways of perceiving computer-mediated work provides a more rigorous basis for the definition and interpretation of similarities and differences among people. These and related questions should be ideal for a merging of the psychological, sociological and anthropological traditions of organizational research, particularly when studying computerization of the workplace.

Practical implications In terms of practical issues, the most obvious implication of our findings concerns the differences in disposition toward computer-mediated work when looking at gender and type of computer used. For managers, the results obtained present new confirmation that computerization of work does affect workers in two countries differently. This should help managers when trying to adapt their company’s computerization of work to a given country. The research also indicates that women’s experiences are different from men’s in two countries. This is even more significant when considering that this study attempted to include respondents of both sexes with similar jobs and responsibilities. It also weakens Gutek and Bikson’s (1985) claim that opportunity structure and status difference in firms may be more important. The results in this study indicate that the women’s experiences need to be taken into careful consideration to avoid possible resistance to computerization (Gattiker and Larwood, 1986b).

The implications of this study’s results are highly complex. An attempt has been made in this research to follow the call that the narrow domestic paradigm be replaced with a more international paradigm by studying respondents from two countries (Adler, 1983). Furthermore, information systems literature suggests that expert systems and decision support systems need to be better integrated (Turban and Watkins, 1986). This study indicated that, even though they may be apparent to an objective outsider, hierarchical differences in computer utilization are not necessarily perceived as such by the employees. Another study found that job-holders usually perceive their work characteristics differently than outside experts who might evaluate them (Znanieck Lopata et al., 1985). Our results indicate two factors relating to employees’ computer attitudes. First, Canadian users feel communication negatively affected by working with main- frame terminals. Second, lower level employees in the U.S. may perceive increased control exercised with such systems. If these two concerns can be addressed by system designers, the productivity increases expected when introducing advanced computer technology in the workplace may very well be reached (Gattiker, 1984; Turban and Watkins, 1986). It is imperative to know individuals’ beliefs about computers in order to achieve maximum effectiveness. The findings of this research project should be of help to managers in realizing this goal.

If we go beyond the simplest personnel issues, our dispositional findings have additional and perhaps more important implications. When we consider, for example, the research findings that depressives are more realistic in their judgments of risk and causation than others (e.g. Alloy and Abramson, 1979), we are led to the conclusion that overly positive dispositions toward computer- mediated work could actually hinder performance in some managerial and staff roles. In contrast to this, however, other research indicates that negative dispositions may lead to computer resistance


(e.g. Gattiker and Larwood, 1986). We may need to examine each organizational role for the amount of enthusiasm versus critical thinking involved in order to best fit jobs with individual’s dispositions toward computer-mediated work. As we pursue the dispositional perspective further, new theory as well as revisions to managerial practice are likely to be suggested.

References Adler, N. J. (1983). ‘Cross-cultural management research: The ostrich and the trend’, f i e Academy of

Alloy, L. B. and Abramson, L. Y. (1979). ‘Judgment of contingency in depressed and nondepressed

Bhagat, R. S. and McQuaid, S. J. (1982). ‘Role of subjective culture in organizations: A review and

Blalock, H. M. (1984). Basic Dilemmas in the Social Sciences, Sage Publications, Beverly Hills. Bodmer, W. (1982). ‘Uberlegungen zur Anwendung der Mikroelektronik. Ergebnisse der Beratungen eines

Gesprachskreises der SZF’, Wirtschaftspolitische Mitteilungen, 38 (5 ) . Brousseau, K. R. (1983). ‘Toward a dynamic model of job-person relationships: Findings, research

questions, and implications for work system design’, Academy of Management Review, 8, 33-45. Cammann, C. (1981). ‘Comments on the organization assessment research program’. In: Van de Ven, A. H.

and Joyce, W. F. (Eds) Perspectives on Organization Design and Behavior, New York, pp. 199-210. Carter, N. M. (1984). ‘Computerization as a predominant technology: Its influence on the structure of

newspaper organizations’, Academy of Management Review, 27,247-270. Child, D. (1970). f i e Essentials of Factor Analysis, Holt, Rinehart & Winston, New York. Demby, E. H. (1985). ‘The new requirement for the researcher: Interactivity with the personal computer’,

Doswell, A. (1983). office Automation, Wiley, New York. Evan, W. M. (1976). Organization Theory: Structures, Systems and Environments, Wiley, New York. Form, W. and McMillen, D. B. (1983). ‘Women, men, and machines’, Work and Occupations, 10, 147-178. Gattiker, U. E. (1984). ‘Managing computer-based office information technology: A process model for

management’. In: Hendrick, H. W. and Brown, O., Jr. (Eds) Human Factors in Organizational Design, Elsevier Science, Amsterdam, The Netherlands, pp. 395403.

Gattiker, U. E. and Coe, L. (1986). ‘Relationship of computer attitudes with perception of career success’. In: Pearce, J. A. I1 and Robinson, R. B. (Eds) Best Paper Proceedings ofthe 46th Annual Meeting of the Academy of Management, 31,294-298.

Gattiker, U. E., Gutek, B. A. and Berger, D. E. (1985). ‘Perceptions of office technologies by employees’. In: Kling, R. (Chair), Computerization in workplaces. Symposium conducted at the 79th annual meeting of the American Sociological Association, Washington, DC.

Gattiker, U. E. and Larwood, L. (1986a). ‘Career success and organizational position: A study of managers and support personnel’, Journal of Business and Psychology, 1,77-87.

Gattiker, U. E. and Larwood, L. (1986b). ‘Resistance to computerization’. In: Gold, B. (Chair), Computerization and Productivity. Symposium conducted at the annual meeting of the TIMS / ORSA, Los Angeles.

Management Review, 8, 226--232.

students: Sadder but wiser?’ Journal of Experimental Psychology: General, 108,441-485.

directions for future research’, Journal of Applied Psychology Monograph, 67,653-685.

European Research, 13,26-28.

Giuliano, V. (1982). ‘The mechanization of work’, Scientific American, September, 149-164. Glick, W. H., Jenkins, G. D. and Gupta, N. (1986). ‘Method versus substance: How strong are underlying

relationships between job characteristics and attitudinal outcomes?’ Academy of Management Journal, 29,441464.

methods of clustering countries on the basis of employee attitudes’, Human Relations, 38,813-840.

Woman, Praeger, New York.

offices’, Sex Roles, 13, 123-136.

I. (Eds) International Review of Industrial and Organizational Psychology, pp. 2 17-234.

Griffeth, R. W., Horn, P. W., DeNisi, A. S. and Kirchner, W. K. (1985). ‘A comparison of different

Gutek, B. A. (1983). ‘Women’s work in the office of the future’. In: Zimmerman, J. (Ed) The Technological

Gutek, B. A. and Bikson, T. K. (1985). ‘Differential experiences of men and women in computerized

Gutek, B. A., Larwood, L. and Stromberg, A. (1986). ‘Women at work’. In: Cooper, C. L. and Robertson,

Hays, W. L. (1981). Statistics, 3rd edn, CBS College Publishing, New York.


Hofstede, G. (1984). Culture 5 Consequences, Sage Publications, Beverly Hills. Kahn, R. L. (1981). ‘Work and health: Some psychosocial effects of advanced technology’, In: Gardell, B.

and Johansson, G. (Eds) Working Life, Wiley, Chichester, U.K., pp. 17-37. Kaiser, H. F. (1974). ‘An index of factorial simplicity’, Psychometrika, 39, 31-36. Karsh, B. and Cole, R. E. (1968). ‘Industrialization and the convergence hypothesis: Some aspects of

contemporary Japan’, Journal of Social Issues, 24,45-63. Kiesler, S., Zubrow, D., Moses, A. and Geller, V. (1985). ’Affect in computer-mediated communication:

An experiment in synchronous terminal-to-terminal discussion’, Human-Computer Interaction, 1, 77- 104.

77-96. Kling, R. and Iacono, S. (1984). ‘Computing as an occasion for social control’, Journal of Social Issues, 40,

Lieberman, M. A., Selig, G. J. and Walsh, J. J. (1982). Office Automation, Wiley, New York. Locke, E. A. (1976). ‘The nature and cause of job satisfaction’. In: Dunnette, M. D. (Ed) Handbook of

Megaw, E. D. and Lloyd, E. J. (Eds) (1984). ‘Human-machine interaction’, Ergonomics Abstracts, Special

Mankin, D., Bikson, T. K. and Gutek, B. A. (1982). ‘The office of the future: Alternatives and choices’,

Morgall, J. (1983). ‘Typing our way to freedom. Is it true that new office technology can liberate women?’

Nunnally, J. C. (1978). Psychometric Eheory, 2nd edn,. McGraw-Hill, New York. O’Reilly, Ch. A., Parlette, G. N. and Bloom J. R. (1980). ‘Perceptual measures of task characteristics: The

biasing effects of differing frames of reference and job attitudes’, Academy of Management Journal, 23,

Industrial and Organization Psychology, Rand McNally, Chicago.

Issue, 16, (1) OOO

The Futurist, 16, 33-37.

Behaviour and Information Technology, 2,215-226.

118-131. Panko, R. R. (1984). ‘Office work’, Office: Technology and People, 2,205-238. Pava, C. (1983). Managing New Office Technology, Free Press, New York. Rousseau, D. M. (1979). ‘Assessment of technology in organizations: Closed versus open systems

Roznowski, M. and H u h , Ch. L. (1985). ‘Influences of functional specialty and job technology on approaches’, Academy of Management Review, 4,53 1-542.

employees’ perceptual and affective responses to their jobs’, Organizational Behavior and Human Decision Processes, 36, 186-208.

Salancik, G. R. and Pfeffer, J. (1977). ‘An examination of need satisfaction models of job attitudes’, Administrative Science Quarterly, 22, 427456.

Salzman, H. and Mirvis, P. (1985). ‘The workforce transition to new computer technologies: Changes in skills and quality of work life’, (Tech. Rep. No. 85-8). Boston University, Center for Applied Social Science, Boston.

Journal of Management, 12, 19-30.

Sociological Review, 48, 824-837.

attitudes’, Journal of Applied Psychology, 70,469480.

longitudinal test’, Administrative Science Quarterly, 31, 56-77.

Psychology Bulletin, 3, 143-158.

Journal of Applied Psychology, 56, 140-145.

Quarterly, 10, 121-136.

Measures in the Social Sciences, 2nd edn, Houghton Mifflin, Boston.

integration, and comments’, Academy of Management Review, 3, 267-280.

Sekaran, U. (1986). ‘Mapping bank employee perceptions of organizational stimuli in two countries’,

Spenner, K. I. (1983). ‘Deciphering Prometheus: Temporal change in the skill level of work’, American

Staw, B. M. and Ross, J. (1985). ‘Stability in the midst of change: A dispositional approach to job

Staw, B. M., Bell, N. E. and Clausen, J . A. (1986). ‘The dispqsitional approach to job attitudes: A lifetime

Triandis, H. C. (1977). ‘Cross-cultural social and personality psychology’, Personality and Social

Triandis, H. C. and Vassiliou, V. (1972). ‘Interpersonal influence and employee selection in two cultures’,

Turban, E. and Watkins, P. R. (1986). ‘Integrating expert systems and decision support systems’, MIS

Webb, E. J., Campbell, D. T., Schwartz, R. D., Sechrest, L. and Grove, J. B. (1981). Nonreactive

White, J. K. (1978). ‘Individual differences and the job quality-worker response relationship: Review,

Weick, K. E. (1979). Ehe Social Psychology of Organizing, 2nd edn, Addison Wesley, Reading, MA. Znanieck Lopata, H., Fordham Norr, D., Barnewolt, D. and Miller, Ch. A. (1985). ‘Job complexity as

perceived by workers and experts’, Work and Occupations, 12, 395415.

Documents

Computerized offices in Canada and the United States: Investigating dispositional similarities and differences