Does ICT use improve contextual performance? Recent evidence from the European Union

Adel BEN YOUSSEF
(University of Nice Sophia Antipolis & GREDEG-CNRS) adel.ben-youssef@unice.fr

Ludivine MARTIN
(CEPS/INSTEAD & CREM-UMR CNRS) ludivine.martin@ceps.lu

Nessrine OMRANI
(University of Paris Sud & ADIS) nessrine.omrani@u-psud.fr

Abstract

The impact of workers’ ICT usages on their excepted performance is a puzzling topic. An optimistic literature has shown the increased role of these technologies inside the organizations. Technology is allowing workers to do bettering and faster work. Economic theory has focused since that its topic on the increase of the productivity and worker’s performance. Several articles have tried to established such a link and have found a significant increase over recent years in the productivity ([Aral et al.2007], [Licht and Moch1999], [Lichtenberg1995], [Greenan and Mairesse2000], [Hans-­‐ Olof Hagén et al.2008]). One challenging view is to consider that ICT uses modify other outcomes within the firm. Performance is not just a result; it also can be an individual behavior that benefits to the organization. Performance can be divided up in terms of task and contextual behaviors. Whereas task performance describes obligatory behaviors, contextual behaviors are behaviors that do not fulfill specific aspects of the job’s required role ([Borman and Motowidlo1997]). A plausible assumption consists of thinking that the behaviors of the workers may become friendlier, cooperatively and sportsmanship thanks to those technologies. These changes benefit the entire firm. In fact, it’s well established that ICT is improving the flexibility, the collaboration and the cooperation inside the firm. Workers are feeling better, becoming more productive and their creativity is stimulated. The opposite view needs also to be considered. In fact, ICT uses may cause “techno-­‐stress” and harm the productivity of the workers. They may have less job satisfaction and lesser performances. Our main hypothesis is that ICT usage is modifying the contextual performance of the workers prior to their output performance (productivity). According to this, ICT are playing a central role in setting cooperative behaviors and inducing workers to be more

1

involved in the production process. These changes are modifying the tasks performances, the contextual performance and also the job satisfaction. The aim of our article is to tackle this point. We want to understand to which extent Internet and Computer uses modify the behaviors of the workers inside the firm. Our research strategy is to consider contextual performance as indicator of the behaviors of the workers. Our paper analyzes the link between Intensity of usage of Computer and Internet and Workers’ Contextual Performance in Europe in 2005. In order to measure contextual performance we follow the methodology developed by Coleman and Borman. Based on the European Working Condition Survey (EWCS 2005)1, we have tried to verify empirically the positive impact of Internet and computer usage on the contextual performance. The relationship between “Intensity of ICT usage” and Workers’ Performance is examined by using an ordered probit model. The model determines the effect of “Internet usage” or “Computer usage” on the probability to get higher contextual performance by a given worker.

Firstly, the study concerns the impact of ICT use on interpersonal citizenship performance, which is defined as behaviors that assist, support, and develop organization members through cooperative and facilitative efforts that go beyond expectations ([Coleman and Borman2000]). H1: ICT use improves workers’ interpersonal citizenship performance. Secondly, the study focuses on the influence of ICT use on organizational citizenship performance, which is defined as citizenship behaviors that demonstrate commitment to the organization through allegiance and loyalty to the organization and organization objectives, and compliance with organizational rules, policies, and procedures ([Coleman and Borman2000]).

1

The survey includes the active population of 27 EU Member States and of 3 other countries (Switzerland, Turkey and Norway) who are aged 16 years and over and the interviews were conducted face-to-face. The basic sample design is a multi-stage, random sampling: in each country, a number of sampling points are drawn with a probability proportional to population size (for a total coverage of the country) and to population density. The target number of interviews is 1000 in every country except Cyprus, Estonia, Luxembourg, Malta and Slovenia, in which it is 600. Weights (used in our analysis) are constructed in order to ensure that the distribution by region, locality size, gender, age, economic activity and occupation is identical to that of the Labor Force Survey (from Eurostat) distribution. As we aim to analyze the impact of ICT use on contextual performance in firms, we excluded self-employees and employee in one-person firms. Thus the sample used focuses on the salaried employees in firms or establishments with two or more employees. We also restricted the sample to the countries that belong to the Euro-zone in 2005 and to Nordic countries, United Kingdom and Switzerland with comparable living and working conditions. The numbers of individuals in the sample is about 11200. A series of questions is devoted to “information and communication technologies” and in particular on computer and Internet use for professional purposes. We also have information on workers behavior used to construct our contextual performance indicators.

2

H2: ICT use improves workers’ organizational citizenship performance. Thirdly, the study aims to analyze the impact of using ICT on job/task conscientiousness, which is defined as extra efforts that go beyond role requirements, demonstrating dedication to the job, persistence, and the desire to maximize one’s own job performance ([Coleman and Borman2000]). H3: ICT use improves workers’ job/task conscientiousness. We use data from the 2005 European Working Condition Survey (EWCS). The aim of the survey is to provide an overview of the state of quality of work and employment in Europe. In order to analyze the impact of ICT use at work on the contextual performance of workers and as we have a lot of information on workers’ behavior to characterize their contextual performance, a data mining technique is used to form groups of workers relative to their behavioral proximity. Built on [Coleman and Borman2000], we distinguish three dimensions of workers’ contextual performance: Interpersonal Citizenship Performance (ICP), Organizational Citizenship Performance (OCP) and Job/Task Conscientiousness (JTC). Ten binary or ordered variables are used to construct the ICP index, twenty for the OCP index and four for de the JTC index. All of the variables used are described in the Table 1 in the Appendix. In order to construct the three indexes, we perform a multiple correspondence analysis followed by a cluster-analysis is performed. The cluster-analysis permits to group individuals in classes that are the most homogeneous as possible according to their similarities with respect to all variables. The classification was based on the coordinates of individuals on the axes we obtain with a multiple correspondence analysis (MCA)2. The hierarchical clustering method we perform uses the Ward index to measure the distance between two classes3. The results of the hierarchical clustering method led to the creation of the three ordered indexes. The ICP index and the OCP index include four groups of workers, and the JTC one, three groups of workers from the lowest ‘performing’ group of worker to the best one. In order to test the impact of ICT use on the three indexes, we estimate ordered probit models characterized by:

yi* = β ' X i + ε i
With yi* the latent utility of the individual i that depends on several independent variables X i and ε i being a random error distributed normally. Here yi* is an unobserved measure of each

2 3

We retain the maximum number of axes that contribute to a large part of the inertia calculated by the MCA. The choice of the number of classes has been determined according to three rules: Je(2)/Je(1) and pseudo Tsquared of [Duda and Hart1973] and pseudo-F of [Calinski and Harabasz1974]. The number of classes retained gives the maximum number attained by each index.

3

dimension of contextual performance, taking values from low y* for which workers are not ‘performant’; to high y* for which workers are very ‘performant’. Our variables of interest are the intensity of computer and of Internet use that takes value from 0 to 44. In all of the three models, we also introduce variables in Xi to characterize workers, firms and countries. We introduce several explanatory variables related to the characteristics of the worker (gender, age, education level, marital status, income band, type of contract, tenure, teleworking, commuting time and occupations), the firm (size and business sector) and the country of residence (variables measured at the country level in order to control for the position in the economic cycle). The descriptive statistics concerning these variables are presented in the Appendix (Table 2). Our results5 show that: 1- Computer and Internet use increase the probability of having Interpersonal Citizenship Performance. 2- Internet use increases the probability of having Organizational Citizenship Performance while computer use has no significant impact on OCP. 3- Internet use also increases Job/Task Performance while computer use does not impact the probability of having JTP.

Key words: contextual performance, ordered probit model, computer usage, Internet usage.

7. References

[Aral et al.2007]Aral, S., Brynjolfsson, E., and Van Alstyne, M. (2007). Productivity effects of information diffusion in email networks. In International Conference on Information Systems. [Borman and Motowidlo1997]Borman, W. C. and Motowidlo, S. J. (1997). Task performance and contextual performance: The meaning for personnel selection research. Human Performance, 10:99–109. [Calinski and Harabasz1974]Calinski, T. and Harabasz, J. (1974). A dentrite method for cluster analysis. Communications in Statistics, 3:1–27.

4

For the intensity of computer use: 0. Never; 1. Less than 1/4 of the time; 2. Less than 3/4 of the time; 3. Almost all of the time; 4. All of the time; for the intensity of Internet use: 0. Never; 1. Almost never; 2. Around 1/4 of the time; 3. Around half of the time or around 3/4 of the time; 4. Almost all of the time or all of the time. 5 See table 3

4

[Coleman and Borman2000]Coleman, V. I. and Borman, W. C. (2000). Investigating the underlying structure of the citizenship performance domain. Human Resource Management Review, 10:25–44. [Duda and Hart1973]Duda, R. and Hart, P. (1973). Pattern classification and scene analysis. New York, Wiley. [Greenan and Mairesse2000]Greenan, N. and Mairesse, J. (2000). Computer and productivity in france: Some evidence. Economics of Innovation and New Technology, 9:275–315. [Hans-­‐Olof Hagén et al.2008]Hans-­‐Olof Hagén, H., Ahlstrand, C., Daniels, M. andNilsson, E., and Adermon, A. (2008). An empirial analysis of innovation 2002-­‐2004 and its impact on productivity. In Statistics Canada Socioeconomic Conference. [Licht and Moch1999] Licht, G. and Moch, D. (1999). Innovation and information technology in services. The Canadian Journal of Economics, 32(2):363–383. [Lichtenberg1995]Lichtenberg, F. (1995). The output contributions of computer equipment and personnel : a firm level analysis. Econimics of innovation and new technology, (3-­‐4).

Appendix

Table 1: Dependent variables ICP (1-4): Interpersonal Citizenship Performance: Behaviors that assist, support, and develop organization members through cooperative and facilitative efforts that go beyond expectations. Clustering methodology on the following items: Colleagues’ Can the worker get assistance from colleagues if he asks for it? assistance (1) Almost never; (2) Rarely; (3) Sometimes; (4) Often; (5) Almost always. Superior' assistance Rotating tasks Team work Pace of work dependent on colleagues' work Political/trade union activity Can choose the order of tasks Can choose the method of work Can choose the speed of work Good friends at work Can the worker get assistance from his superior if he asks for it? (1) Almost never; (2) Rarely; (3) Sometimes; (4) Often; (5) Almost always. Equal 1 when the job involves rotating tasks between the employee and colleagues, 0 otherwise. Equal 1 when the job involves doing all or part of the work in a team, 0 otherwise. Equal 1 when the pace of work is dependent on the work done by colleagues, 0 otherwise. How often is the worker involved in political or trade union activity? (1) Never; (2) Once or twice a year; (3) Once or twice a month; (4) More often Equal 1 when the worker is able to choose or change the order of tasks, 0 otherwise. Equal 1 when the worker is able to choose or change the methods of work, 0 otherwise. Equal 1 when the worker is able to choose or change the speed or rate of work, 0 otherwise

Does the worker has very good friends at work? (1) Strongly disagree; (2) Disagree; (3) Neither agree nor disagree; (4) Agree; (5) Strongly agree. OCP (1-4): Organizational Citizenship Performance: citizenship behaviors that demonstrate commitment to the organization through allegiance and loyalty to the organizationand organization objectives, and compliance with organizational rules, policies,and procedures.

5

Clustering methodology on the following items: Meeting precise Equal 1 when the main paid job involves meeting precise quality standards, 0 quality standards otherwise. Job security Well paid for the work Good prospects for career advancement Satisfaction Information about health and safety risks Work/life balance Opportunity to do what you do best Being 'at home' in the organisation Might the worker lose his job in the next 6 months? (1) Agree or strongly agree; (2) Neither agree nor disagree; (3) Disagree; (4) Strongly disagree. Is the worker well paid for the work he does? (1) Strongly disagree; (2) Disagree; (3) Neither agree nor disagree; (4) Agree; (5) Strongly agree. Does the job offers good prospects for career advancement? (1) Strongly disagree; (2) Disagree; (3) Neither agree nor disagree; (4) Agree; (5) Strongly agree. On the whole, is the worker satisfied with the working conditions in his main paid job? (1) Not at all satisfied; (2) Not very satisfied; (3) Satisfied; (4) Very satisfied. Regarding the health and safety risks related to performance of the job, how well informed would the worker say he is? (1) Not at all well informed; (2) Not very well informed; (3) Well informed; (4) Very well informed. Does the working hours fit in with the family or social commitments outside work of the worker? (1) not at all well or not very well, (2) well or (3) very well. At work, does the worker has the opportunity to do what he does the best? (1) Almost never or rarely; (2) Sometimes; (3) Often; (4) Almost always. Does the worker feel itself 'at home' in his organisation? (1) Strongly disagree; (2) Disagree; (3) Neither agree nor disagree; (4) Agree; (5) Strongly agree. At work, does the worker has opportunities to learn and grow? (1) Strongly disagree; (2) Disagree; (3) Neither agree nor disagree; (4) Agree; (5) Strongly agree. Equal 1 when the main paid job involves learning new things, 0 otherwise. Can the worker takes his break when he wish? (1) Almost never; (2) Rarely; (3) Sometimes; (4) Often; (5) Almost always. Has the worker the feeling of doing useful work? (1) Almost never; (2) Rarely; (3) Sometimes; (4) Often; (5) Almost always. Does the worker has to interrupt a task he is doing in order to take on a non disruptive task? (0) Never; (1) Interruption but disruptive; (2) Occasionally; (3) Fairly often; (4) Very often Is the worker able to apply his own ideas in his work? (1) Almost never or rarely; (2) Sometimes; (3) Often; (4) Almost always. In the past twelve months, has the worker been contacted by email or telephone, in matters concerning his main paid job outside his normal working hours? (0) Never; (1) Less often; (2) A couple of times a month; (3) At least once a week; (4) Every day. Equal 1 when the main paid job involves to assess alone the quality of own work, 0 otherwise. Equal 1 when the main paid job involves solving unforeseen problems on your own, 0 otherwise. Equal 1 when over the past 12 months the worker has discussed work-related problems with an employee representative, 0 otherwise.

Learn and grow

Learning new things Taking breaks when you wish Doing useful work Non disruptive tasks Apply own idea Frequency of contact by email or telephone outside working hours Alone assessment of own work quality Solving unforeseen problems alone Discussion on work-related

6

problems with an employee representative Worker consultation Training paid by the employeur

Equal 1 when over the past 12 months, the worker has been consulted about changes in the organisation of work and/or working conditions, 0 otherwise.

Equal 1 when over the past 12 months, the worker has undergone training paid for or provided by the employer to improve his or her skills or on-the-job training (co-workers, supervisors) to improve his or her skills, 0 otherwise. JTC (1-3): Job/Task Conscientiousness: Extra efforts that go beyond role requirements;demonstrating dedication to the job, persistence, and the desire to maximize one's own job performance. Clustering methodology on the following items: Working at home, Does the main paid job involve working at home (excluding telework)? excluding telework (1) Never; (2) Almost never; (3) Around 1/4 of the time or more. Extra hours Training paid alone Self training Equal 1 when the worker entirely determined alone his working hours and work more than the legal time observed in the country, 0 otherwise. Equal 1 when over the past 12 months, the worker has paid a training to improve his or her skills, 0 otherwise. Equal 1 when over the past 12 months, the worker has undergone a training like self-learning or on-line tutorials, 0 otherwise.

Table 2: Descriptive statistics Mean Contextual performance ICP ICP 1 2 3 4 OCP OCP 1 2 3 4 JTC JTC 1 2 3 e-skills Frequency of Internet use (0-4) Frequency of Internet use (0-4) 0 1 2 3 4 Frequency of computer use (0-4) 58.52% 20.57% 20.91% 1.4487 46.22% 11.80% 13.07% 8.70% 20.21% 1.5573 0.4927 0.4042 0.4067 1.5995 0.4986 0.3227 0.3371 0.2818 0.4016 1.5611 0 4 8.84% 38.08% 40.32% 12.77% 1.6239 0.2839 0.4856 0.4906 0.3337 0.8080 36.14% 19.88% 21.77% 22.21% 2.5701 0.4804 0.3991 0.4127 0.4156 0.8230 2.3004 Std. Dev. 1.1735 Min 1 Max 4

1

4

1

3

0

4

7

Frequency of computer use (0-4) 0 1 2 3 4 Worker's characteristics Male Age Age squared/1000 Education level (Ref. No or primary education) Lower secondary education High secondary Post secondary Tertiary education Living with a partner or a spouse Income band Indefinite contract Full time job Tenure Tenure squared/100 Teleworking Commuting time (min) Occupations (Ref. Service workers and sales workers) Legislators and managers Professionals Technicians and associate professionals Clerks Craft and related trades workers and skilled agricultural and fishery workers Plant and machine operators and assemblers Elementary occupations Firm's characteristics Size of the firm (Ref. 2-9 employees) 10-49 employees 50-249 employees 250+ employees Business sectors (Ref. Agriculture, mining, energy) Manufacturing Construction Trade Transports and telecom. Finance and insurance Public administration, education, health Services Country's characteristics Macroeconomic control variables Country unemployment rate Country percentage of people with at least secondary education Country growth rate Number of observations Weighted statistics 37.40% 20.49% 12.15% 8.88% 21.08% 54.33% 41.1491 1.8137 16.64% 42.68% 7.34% 27.80% 26.94% 5.9990 84.47% 82.94% 11.2022 2.0945 0.2246 21.5727 6.46% 16.37% 17.02% 14.65% 13.98% 8.89% 9.88% 0.4839 0.4037 0.3267 0.2844 0.4079 0.4981 10.9768 0.9162 0.3725 0.4946 0.2608 0.4481 0.4437 2.8984 0.3622 0.3761 9.1634 3.1814 0.5643 16.2040 0.2459 0.3700 0.3759 0.3536 0.3468 0.2846 0.2984 0 16 1 76

0 0 0 0 0 1 0 0 2 0 0 0 0 0 0 0 0 0

1 1 1 1 1 10 1 1 50 2 150 1 1 1 1 1 1 1

34.01% 25.06% 19.05% 20.43% 7.14% 16.62% 6.66% 4.13% 14.72% 26.18%

0.4738 0.4334 0.3927 0.4032 0.2575 0.3722 0.2494 0.1991 0.3543 0.4396

0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1

8.03% 69.26% 1.79% 11196

2.3339 14.1973 1.0165

4.4 26.5 0.7

10.7 88.2 6.2

8

Table 3: Ordered probit regression ICP (1) 0.0275** (0,013) 0.0311*** (0,011) 0,0418 (0,037) -0.0212** (0,010) 0.219** (0,110) -0,0215 (0,096) 0,0139 (0,113) 0,114 (0,098) -0,0543 (0,120) -0,000589 (0,052) -0,0101 (0,013) -0,0232 (0,050) -0,0152 (0,027) 0,00452 (0,007) 0,0107 (0,018) -0.0774** (0,038) 0,00155 (0,002) 0,159 (0,131) 0,0673 (0,087) 0.107** (0,052) 0,0138 (0,060) 0,0175 (0,040) 0,0251 (0,078) -0,0943 (0,082) 9 OCP (2) 0.118*** (0,021) 0,0191 (0,024) 0,0186 (0,042) -0.0288*** (0,010) 0.318*** (0,110) 0,00878 (0,056) 0.130** (0,065) 0.222** (0,103) 0.198** (0,101) -0,00348 (0,047) 0.0358*** (0,005) -0,00747 (0,055) 0.0669** (0,033) 0,00565 (0,007) 0,00717 (0,018) 0.0414*** (0,011) -0,00271 (0,002) 0.156*** (0,053) 0.212** (0,090) 0.221*** (0,074) -0,0838 (0,054) 0,0329 (0,065) -0.167* (0,102) -0,16 (0,100) JTC (3) 0.0881*** (0,020) -0,00295 (0,025) 0,0441 (0,048) -0,00612 (0,009) 0,0413 (0,123) 0.132* (0,080) 0,11 (0,145) 0.267* (0,145) 0.429*** (0,148) -0,0419 (0,060) 0.0371*** (0,012) -0.165* (0,099) 0.135* (0,081) -0,00298 (0,006) 0,00685 (0,015) 0.414*** (0,068) 0,000392 (0,001) 0,212 (0,139) 0.363*** (0,085) 0.230*** (0,051) 0,0839 (0,071) -0,0657 (0,072) -0,0885 (0,118) -0.202** (0,102)

Frequency of Internet use (0-4) Frequency of computer use (0-4) Male Age Age squared/1000 Lower secondary education (Ref. No or primary education) High secondary Post secondary Tertiary Living with a partner or a spouse Income band (1-10) Indefinite contract Full time job Tenure Tenure squared/100 Teleworking Commuting time (min.) Legislators and managers (Ref. Service workers and sales workers) Professionals Technicians and associate professionals Clerks Craft and related trades workers and skilled agricultural and fishery workers Plant and machine operators and assemblers Elementary occupations

10-49 employees (Ref. 2-9 employees) 50-249 employees 250+ employees Manufacturing (Ref. Agriculture, mining, energy) Construction Trade Transports and telecom. Finance and insurance Services Public administration, education, health Country unemployment rate (2005) Country percentage of people with at least secondary education (2005) Country growth rate (2005) Cut 1 Cut 2 Cut 3 Observations Log Lik Pseudo R-squared

-0,0687 (0,063) -0.105* (0,060) -0,0486 (0,074) 0,0582 (0,100) 0,173 (0,117) 0,027 (0,043) -0.176*** (0,064) 0,136 (0,110) -0,0122 (0,084) 0,0594 (0,057) -0.0640*** (0,020) 0.00896*** (0,003) 0.132*** (0,044) -0,388 (0,316) 0,14 (0,322) 0.784** (0,345) 11196 -14722 0,0296

-0,00845 (0,079) 0,0476 (0,083) 0.0790** (0,037) -0.184** (0,092) 0,0164 (0,131) -0,0634 (0,095) -0.137* (0,073) 0,0894 (0,109) -0,00395 (0,086) -0,0572 (0,081) -0.0694*** (0,013) 0.00540* (0,003) 0,0427 (0,035) -1.580*** (0,376) -0,172 (0,406) 1.177*** (0,412) 11196 -12600 0,0707

-0,0428 (0,056) -0,0497 (0,111) 0,0694 (0,107) -0,133 (0,087) -0,13 (0,107) -0,0206 (0,102) -0,0064 (0,104) -0,0817 (0,090) -0,0531 (0,146) 0.226** (0,100) -0.0559*** (0,018) 0.00376* (0,002) -0,0389 (0,035) 0.529* (0,271) 1.235*** (0,273) 11196 -9471 0,124

10