Loading...
Please wait, while we are loading the content...
Similar Documents
Information Technology and Organizational Practices: an Exploratory Empirical Analysis
| Content Provider | Semantic Scholar |
|---|---|
| Author | Resende, Marcelo |
| Copyright Year | 2004 |
| Abstract | Rapid price declines are likely to facilitate the increasing diffusion of information technology-IT in industrial firms. Recent contributions by Bresnahan et al. (2002), however, has emphasized, however, the interdependence between IT adoption and workplace organization. The present paper undertakes an exploratory analysis on the complementarities between IT adoption and the utilization of modern organizational practices. The analysis considers 3-digits sectoral data on the Brazilian manufacturing industry in 1996. The multivariate statistical analysis of canonical correlation analysis allowed to investigate the overall relationship between a group of variables reflecting IT adoption (comprising the number of microcomputers per employee (median value), percentage of firms providing computer training for employees, percentage of firms with Internet connection and percentage of firms with intranet (internal data communication network)) and other group of variables reflecting the adoption of modern organizational practices (comprising the percentage of firms adopting total quality management, statistical control of processes, internal just-intime and improvement groups (Kaizen)). The results indicated a significant association between the two groups of variables and the individual variables, in the majority of the cases, displayed positive and significant correlations within their group. The evidence appears to support the perception that the adoption of IT is favored in an environment characterized by the utilization of modern organizational practices. Key-words: Information technology, organizational practices, canonical correlation JEL classification: L23, L86 * The author acknowledges research assistance from Alice Pinho, Pedro Mendonça and Arthur Jóia, and comments from a participant of the European Applied Business Research Conference-EABR (Edinburgh, 2004) on a previous version of this paper but the usual disclaimer applies. Financial support from CNPq is gratefully acknowledged. Introduction The utilization of information technology (IT) and different forms of modern workplace organizational practices have become increasingly widespread and has underscored the importance of identifying the skill content in different production processes. In fact, the high wage dispersion in the U.S. embodies a component that expresses a leading role for high-skilled workers and that exhibits complementarities with supporting activities like training programs and different modalities of workplace organization. In fact, Bresnahan et al. (2002) investigated the interrelationships between IT, workplace organization and new product and services in establishing a skill-biased technical change with important implications in terms of labor demand patterns. Under a similar motivation, Autor et al. (2003) detected that a substantial part of estimated relative labor demand shifts between routine manual and cognitive and non-routine labor in the U.S. The detection of complementarities among those three components warrant further investigations. The introduction of IT does not constitute an isolated initiative but rather is associated with the complementary emergence of an organizational (and intangible) capital. These capabilities are likely to consolidate after trial and error attempts on the introduction of modern organizational practices. IT introduction in principle cannot be dissociated from the gradual organizational redesign of the firm. The present paper intends to explore interrelationhips and complementarities among IT and organizational practice in the Brazilian manufacturing industry. In fact, the diffusion of those modernization efforts is more recent in this large and heterogeneous developing country. For that purpose, Multivariate Statistical Analysis will be useful in order to provide an informative and exploratory characterization of the aforementioned relationship and fully explore its multidimensional features. The application of the related methods in the context of Industrial Economics is relatively scarce. Exceptions are provided by Resende (1995) who uses principal components analysis to check for redundancies among different concentration indexes as applied to U.S. industry, Hollenstein (1996) who considers factor analysis to conceive a composite indicator of innovation for manufacturing firms in Switzerland. The technique was also considered by Kleinknecht et al. (2002) to explore the presence of common dimensions in a set of different innovation indicators in Netherlands. Lima et al (2004), on the other hand, investigated the association between skill-enhancement efforts and performance for firms in the Brazilian chemical sector taking as reference the initial status with respect to quality certification. The technical procedures utilized in the present paper are similar to that paper, but of course consisting of a very distinct context. The paper is organized as follows. The second section presents a brief introduction to the technique of Canonical Correlation Analysis. The third section discusses the data used and presents the empirical results. The fourth section brings some final comments. 2. Canonical Correlation Analysis: Basic Background Multivariate Statistical Analysis constitutes a useful exploratory tool for investigating phenomena characterized by multiple dimensions. In fact, organizational practices and information technology content are likely to comprise multiple and complex interrelating dimensions. The technique of Canonical Correlation Analysis (CCA), in particular, may provide a useful method for assessing linear relationships between groups of variables. It can be thought, to a certain extent, as a generalization of linear multiple regression methods [see Manly (1994) for an introduction]. Consider 2 groups of variables of interest, such that one has p variables X1, X2, ., Xp and the other has q variables Y1, Y2, ., Yq. One can then define linear combinations of the original variables as follows: U1 = a11 X1 + a12 X2 + . + a1p Xp U2 = a21 X1 + a22 X2 + . + a2p Xp . . Ur = ar1 X1 + ar2 X2 + . + arp Xp and V1 = b11 Y1 + b12 Y2 + . + b1q Yq V2 = b21 Y1 + b22 Y2 + . + b2q Yq . . Vr = br1 Y1 + br2 Y2 + . + brq Yq CCA considers r pairs of canonical variates (where the number is defined in terms of the minimum between p and q) such that the correlation between U1 and V1 is maximum whereas for posterior pairs (Ui, Vi) correlation is maximum subject to the restriction of no correlation with previous considered pairs. CCA is somewhat analogous with principal components analysis but the reference criterion relates to correlation and not variance as the latter. A second difference refers to the interpretation of the loadings indicated in the previous equations. The signs of the those coefficients can lead to misleading interpretations given specific patterns of correlation between the variables of the two groups. Possible guidelines are suggested by Alpert and Peterson (1972) and Manly (1994). For example:: i) Consider the correlation between the canonical variate and the original variables of the related group; ii) Consider a graph on the association between the first canonical variates of the different groups (U1 and V1) so as to detect strong idiosyncratic patterns on individual behavior. These strategies are pursued in sub-section 3.2 where the results from the CCA are presented. 3. Empirical Analysis 3.1Data Source The basic data source of the present study was a unique data set generated from an extensive survey conducted in 1996 by the statistical bureau of the state of São Paulo-Brazil [Pesquisa da Atividade Econômica Paulista-PAEP, Fundação SEADE[. The survey collected not only some basic accounting data but also a broad range technological related variables dealing with innovation aspects and yet variables reflecting different aspects of modern organizational practices. The referred information was available for larger firms (with 100 or more employees) and the used sample comprises 2153 firms in 89 3-digits. The exploratory canonical correlation undertaken in the present study consider sectoral-level variables and indicates either the sector ́s median value (in the case of continuous variables) of the percentage of firms adopting a particular practice (in the case of original binary firm-level variables). Specifically, we consider the variables described in terms of two categories: Information technology variables: . NMICRO: number of microcomputers per employee (median value of the sector); . CTRAIN: percentage of firms providing computer training for employees linked to production; . INTER: percentage of firms with Internet access; 1 A similar analysis was pursued in Lima et al. (2004) to investigate the relationship between 2 It is reassuring that the analysis undertaken at the 4-digits level of aggregation produced qualitatively similar results, but the choice of the 3-digits aggregation reflects the intention of assuring representatibe sectors with a good number of firms. . INTRA: percentage of firms with intranet systems (data transmission, for example in terms of LAN networks); Organizational practices variables: . TQM: percentage of firms adopting total quality management; . SCP: percentage of firms adopting statistical control of processes; . IJIT: percentage of firms adopting internal just-in-time procedures; . KAIZEN: percentage of firms adopting improvement groups practices; The table in the appendix displays the related values for the 89 (3-digits) sectors for which data was available. The PAEP survey focuses on manufacturing industry in the state of São Paulo, which is in fact the dominant industrial state in Brazil. It is important to stress that descriptive assessment of IT and modern organizational practices diffusion are relatively scarce. An exception is provided by Forman et al (2002) who conducted a detailed geographic and sectoral assessment of IT di |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://www.ie.ufrj.br/oldroot/publicacoes/discussao/information_technology_and_organizational_practices_an_exploratory_sectorial_analysis.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
