Abstract
This study explores the factors that can impact individual performance when using enterprise resource planning (ERP) systems. Starting from the proposition that organizational performance depends on individuals' task accomplishments, we test a structural model of task–technology fit, ERP user satisfaction, and individual performance in ERP environments. This research utilizes a survey method to examine the perceptions of ERP users. We performed factor and reliability analyses to assess the validity of the survey instrument. Six factors were identified as having an impact on individual performance: System Quality, Documentation, Ease of use, Reliability, Authorization, and Utilization. To explore the relationships among these factors, we conducted regression and multivariate adaptive regression splines analysis, and compared the findings from these two analytical techniques. The study provides evidence that System Quality, Utilization, and Ease of Use are the most important factors bearing on individual performance in ERP environments. Our findings also provide IT managers and researchers with knowledge of how these factors can be manipulated to improve individual performance when using ERP systems.
Similar content being viewed by others
Notes
The survey approach was chosen because we wanted to extend the existing models that were developed via the survey methodology. Other research methods (e.g. qualitative approaches), while valid, would have been: (1) inefficient due to the range of organizations in our study and (2) would have had limited immediate impact on extending the existing models because they would require another step (the survey method) to collect data in order to validate the proposed model.
References
Abraham A (2002) Analysis of hybrid soft and hard computing techniques for Forex monitoring systems. World Congress on Computational Intelligence, pp 1616–1622.
Abraham A and Steinberg D (2001) Is neural network a reliable forecaster on earth? A MARS query! In International Work – Conference on Artificial and Natural Neural Network (MIRA J and PRIETO A, Eds), pp 679–686, Springer-Verlag, Grenada, Spain.
Al-Gahtani SS and King M (1999) Attitudes, satisfaction and usage: factors contributing to each in the acceptance of information technology. Behaviour & Information Technology 18 (4), 277–297.
Al-Mudimigh A, Zairi M and Al-Mashari M (2001) ERP software implementation: an integrative framework. European Journal of Information System 10, 216–226.
Bailey JE and Pearson SW (1983) Development of a tool for measuring and analyzing computer user satisfaction. Management Science 29 (5), 530–545.
Baron R and Kenny DA (1986) The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology 51 (6), 1173–1182.
Bauer E and Kohavi R (1999) An empirical comparison of voting classification algorithms: bagging, boosting, and variants. Machine Learning 36, 105–139.
Bingi P, Sharma M and Godla J (1999) Critical issues affecting an ERP implementation. Information Systems Management 16 (3), 7–14.
Blili SRL and Rivard S (1998) Impact of task uncertainty, end-user involvement, and competence on the success of end-user computing. Information and Management 33 (3), 137–153.
Briand LC, Melo WL and Wust J (2002) Assessing the applicability of fault-proneness models across object-oriented software projects. IEEE Transactions on Software Engineering 28 (7), 706–720.
Carey WP and Yee SS (1992) Calibration of nonlinear solid-state sensor arrays using multivariate regression techniques. Sensors and Actuators 9, 113–122.
Chen L (1997) Multivariate regression splines. Computational Statistics & Data Analysis 26, 71–82.
Delone WH and Mclean ER (1992) Information systems success: the quest for the dependent variable. Information Systems Research 3 (1), 60–95.
Dishaw MT and Strong DM (1998) Assessing software maintenance tool utilization using task–technology fit and fitness-for use models. Journal of Software Maintenance Research and Practice 10 (3), 151–179.
Dishaw MT and Strong DM (1999) Extending the technology acceptance model with task–technology fit constructs. Information and Management 36 (1), 9–21.
Doll WJ and Torkzadeh G (1988) The measurement of end-user computing satisfaction. MIS Quarterly 12 (1), 259–274.
Doll WJ, Xia W and Torkzadeh G (1994) A confirmatory factor analysis of the end-user computing satisfaction instrument. MIS Quarterly 18 (4), 453–461.
Downing CE (1999) System usage behavior as a proxy for user satisfaction: an empirical investigation. Information and Management 35 (4), 203–216.
Ekman T and Kubin G (1999) Nonlinear prediction of mobile radio channels: measurements and MARS model designs. IEEE International Conference on Acoustics, Speech and Signal Processing, pp 2667–2670, Phoenix, Arizona.
Etezadi-Amoli J and Farhoomand AF (1996) A Structural model of end user computing satisfaction and user performance. Information and Management 30 (2), 65–73.
Ferratt TW and Vlahos GE (1998) An investigation of task–technology fit for managers in Greece and the US. European Journal of Information Systems: An Official Journal of the Operational Research Society 7 (2), 123–136.
Friedman JH (1991) Multivariate adaptive regression splines. The Annals of Statistics 19 (1), 1–67.
Goodhue DL (1995) Understanding user evaluations of information systems. Management Science 41 (12), 1827–1844.
Goodhue DL (1998) Development and measurement validity of a task–technology fit instrument for user evaluations of information systems. Decision Sciences 29 (1), 105–138.
Goodhue DL, Klein B and March ST (2000) User evaluations of IS as surrogates for object performance. Information and Management 38 (2), 87–101.
Goodhue DL and Thompson RL (1995) Task–technology fit and individual performance. MIS Quarterly 19 (2), 213–236.
Griffin W, Fisher N, Friedman J and Ryan C (1997) Statistical technique for the classification of chromites in diamond exploration samples. Journal of Geochemical Exploration 59, 233–249.
Hastie TJ and Tibshirani R (1990) Generalized Additive Model. Chapman & Hall, London.
Hastie TH, Tibshirani R and Friedman J (2001) The Elements of Statistical Learning: Datamining, Inference, and Prediction. Springer-Verlag, New York.
Hendrickson AR, Glorfeld K and Cronan TP (1994) On the repeated test-retest reliability of the end-user computing satisfaction instrument: a comment. Decision Science 25 (4), 655–667.
Igbaria M and Tan M (1997) The consequence of information technology acceptance on subsequent individual performance. Information and Management 32 (3), 113–121.
Ives B, Olson MH and Baroudi JJ (1983) The measurement of user information satisfaction. Communications of the ACM 26 (10), 785–793.
Jarvenpaa SL (1989) The effect of task demands and graphical format on information processing strategies. Management Science 35 (3), 285–303.
Jin R, Chen W and Simpson T (2001) Comparative studies of metamodelling techniques under multiple modelling criteria. Journal of Structural and Multidisciplinary Optimization 23 (1), 1–13.
Ko M and Osei-Bryson K-M (2004) Using Regression Splines to Assess the Impact of Information Technology Investments on Productivity in the Health Care Industry. Information Systems Journal 14 (1), 43–63.
Kremers M and Dissel HV (2000) ERP system migrations. Communications of the ACM 43 (4), 53–56.
Kuhnert P, Kim-Anh D and Mcclure R (2000) Combining non-parametric models with logistic regression: an application to motor vehicle injury data. Computational Statistics & Data Analysis 34, 371–386.
Lai VS (1999) A contingency examination of CASE-task fit on software developer's performance. European Journal of Information Systems 8 (1), 27–39.
Mabert V, Soni A and Venkataraman MA (2001) Enterprise resource planning: measuring value. Production and Inventory Management Journal, Summer-Fall 44 (3–4), 46–52.
Mallick BK, Denison DGT and Smith AFM (1997) Bayesian survival analysis using a MARS model. Technical report, Imperial College, London.
Markus LM and Tanis C (2000) The enterprise systems experience – from adoption to success. In Framing the domains of it research: glimpsing the future through the past (ZMUD RW, Ed), Pinnaflex educational resources, Cincinnati, Ohio.
Mathieson K and Keil M (1998) Beyond the interface: ease of use and task–technology fit. Information and Management 34 (4), 221–230.
Morgan G (1986) Images of Organization. Sage Publications, Beverly Hills, CA.
Murray MG and Coffin GW (2001) A case study analysis of factors for success in ERP system implementations. Proceedings of the Seventh American Conference on Information Systems, pp 1012–1018.
Nguyen-Cong V, Van Dang G and Rode DBM (1996) Using multivariate adaptive regression splines to OSAR studies of dihydroartemisinin derivatives. European Journal of Medical Chemistry 31, 797–803.
Palaniswamy R and Frank T (2000) Enhancing manufacturing performance with ERP systems. Information Systems Management 17 (3), 43–55.
Parr A and Shanks G (2000) A model of ERP project implementation. Journal of Information Technology 15, 289–303.
Summer M (1999) Critical success factors in enterprise wide information management systems projects. Proceedings of the 1999 ACM SIGCPR Conference on Computer Personnel Research, pp 297–303.
Thompson RL, Higgins CA and Howell JM (1991) Personal computing: toward a conceptual model of utilization. MIS Quarterly 15 (1), 125–143.
Torkzadeh G and Doll WJ (1991) Test-retest reliability of the end-user computing satisfaction instrument. Decision Science 22 (1), 26–37.
Wilson EV and Addo TBA (1994) An investigation of the relative presentation efficiency of computer-displayed graphs. Information and Management 26 (2), 105–115.
York TP and Eaves LJ (2001) Common disease analysis using multivariate adaptive regression splines (MARS): genetic analysis workshop 12 simulated sequence data. Genetic Epidemiology 21, 649–654.
Acknowledgements
This research was supported in part by a grant from the 2004 Summer Research Program of the School of Business of Virginia Commonwealth University, Richmond, VA, U.S.A.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix A
Appendix B: Overview of RS
RS is a method for flexible regression modeling (Friedman, 1991). While traditional regression analysis uses a single slope to represent the relationship between a dependent and an independent variable, RS allows more than one slope, thus it provides a more flexible model than regression analysis. The RS method uses a set of piecewise polynomials to describe the relationships between the independent variables and the dependent variable (Chen, 1997). Each piecewise polynomial, called a basis function (BF), represents a sub-region of the domain of the spline. Basis functions are used by RS to implement the concepts of knots and piecewise linear RS (Briand et al., 2002).
MARS is a computer-assisted approach to RS analysis that allows for automatic selection of the knots (Hastie & Tibshirani, 1990; Friedman, 1991). MARS builds the RS model in a two-phase process, using a forward-stepwise regression selection and backward-stepwise deletion strategy. In the first phase, an overfitted model is built by adding basis functions. In the second phase, basis functions that have the least contribution to the model are deleted where removal causes the smallest increase in residual squared error. The MARS model is optimized based on the Generalized Cross Validation measure (Hastie et al., 2001). A MARS model can be generated that allows no interaction between the input variables, or to permit interactions between two or more variables. In MARS, analysis interactions between variables have a hierarchical, tree-like structure, with parent and child relationships between basis functions. A MARS model that does not involve interaction between the variables (or factors) can be expressed as a linear combination of nonlinear functions (i.e., BF jk ) of the independent variables such that y=b0+Σ j f j (x j ). Equivalently, it can be expressed as a linear combination of basis functions BF jk such that y=β0+Σ j Σk∈Kj β jk BF jk where f j (x j )=Σk∈Kj β jk BF jk , β0 is a constant that is equivalent to the intercept in the regression model, BF jk is a basis function of the independent variable (or factor) that has one of the following forms: max(0, X j −c jk ), or max(0, c jk −X j ), and β jk is a coefficient of the basis function BF jk . MARS must both identify the most relevant knots (e.g. c jk ) as well as the coefficients. The coefficients are the coefficients β jk and are estimated by minimizing the sum of square errors. RS performs a polynomial fit in each region with constraints at the knots using the least squares criterion.
Appendix C: Details of factor analysis
Independent variables:
We conducted several iterations of factor analyses in order to find a meaningful or interpretable grouping of the questionnaire items. Based on the results of the initial iteration, RDATA, TRAIN, FUNC2, FUNC3, FUNC4, FUNC5, SUPP1, and SUPP2 variables were omitted from further consideration because they have cross loadings (i.e., the load was very close on more than one factors and the difference in the loadings was less than 0.1) and low communalities. Although the difference in the loadings of MEAN on the System Quality factor (i.e., factor 1) and the Ease of Use factor (i.e., factor 3) was less than 0.1 (i.e., loading on factor 1 and 3 were 0.607 and 0.519, respectively), it was retained for the following reasons: (1) the loading on the System Quality factor was 0.088 higher than the loading on the Ease of Use factor; (2) the alpha of System Quality with MEAN was as high as 0.9603; and (3) the reliability analysis of the System Quality factor did not suggest that MEAN should be deleted.
In our second iteration, factor analysis and reliability analysis were re-performed on the other 26 items (cf. Table C1). The five factors produced by the factor analysis explained 70.996% of the total variance. Reliability analysis was then performed on each of the factors suggested by the factor analysis. Although the reliability analysis recommended that RDETAIL should be dropped, the item was maintained because the alpha of System Quality is quite high (0.9603) and if the item is deleted, the increase in the alpha is not significant (0.0002).
The exploratory factor analysis confirmed the overlap of the instruments, as described by Goodhue (1998). The questionnaire items regarding Ease of Use from the TTF instrument and from the user satisfaction instrument of Etezadi-Amoli and Farhoomand were loaded together into one factor (Ease of Use). In addition, the currency item from the TTF instrument and the timeliness item from the user satisfaction instrument of Etezadi-Amoli and Farhoomand were loaded on the System Quality factor.
Dependent Variables:
Utilization and Individual Performance are dependent variables in this study. However, Utilization is also an independent variable for Individual Performance. Thus, factor analyses were performed separately on Individual Performance items and Utilization items.
Items from Individual Performance (i.e., PERFO1, PERFO2, PERFO3) were loaded together into one factor with 81.79% total variance explained. The reliability analysis recommended the item PERFO3 to be deleted. However, it was maintained since the alpha with the three items was as high as 0.927, and the alpha would be increased by only 0.02, if PERFO3 is deleted. Items from Utilization were loaded together into one factor with 28% total variance explained. Results of factor analysis and reliability analysis are presented in Tables C2 and C3.
Rights and permissions
About this article
Cite this article
Kositanurit, B., Ngwenyama, O. & Osei-Bryson, KM. An exploration of factors that impact individual performance in an ERP environment: an analysis using multiple analytical techniques. Eur J Inf Syst 15, 556–568 (2006). https://doi.org/10.1057/palgrave.ejis.3000654
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1057/palgrave.ejis.3000654