Innovative Behavior of IT Services Firms in Portugal and Denmark Luísa Ferreira Lopes DIMETIC Session, Maastricht 8-19 October 2007
Research Questions What different innovation profiles (patterns of innovative behavior) can be identified in IT services firms ? Can we find alternative ways of measuring innovation intensity in services that may be more reliable than existing measures ?
Motivation Why Services ? Why IT ? Why innovation assessment ?
Litterature Review Effects of IT services as a source of innovation, elsewhere in the economy Innovation activity within IT services Torrisi (1998) Howells (2000) Mamede (2002) Weterings (2006)
Methodology Process of collecting data: semi-structured face-to-face interviews Questions format: mostly closed and some opened
Methodology Questionaire: 11 sections General information Markets Supply Innovation process Innovation output Innovation input Innovation impact/effects Conditioning factors of innovation Management characteristics Human resources Networking
Methodology Data Collection No reference was made to innovation before the interview 31 interviews in Denmark 31 interviews in Portugal With CEO (except 4 firms) Most frequent duration 1h30m
Results – Cluster Analysis 258 vars 72 vars 12 vars 6 vars Trigger factors Export to developed countries Market scope Innovation importance Innovation intensity Competitive position
Results – Cluster Analysis 2 clusters: “Active firms” N=45 (72,6%) Internal innovation trigger factors Export to developed countries Larger market scope Innovation more important Innovate more intensively Consider they have a better competitive position “Passive firms” N=17 (27,4%) The symetrical
Results – Discriminant Analysis Examine whether firms in the two clusters can be distinguished from each other based on a linear combination of variables Similar process for selecting the variabels 10 variables
Results – Discriminant Analysis All statistical tests indicate a high quality of the discriminant model Classification results: 91.9 % original firms correctly classified 90.3 % cross-validated firms correctly classified
Results – Discriminant Analysis Discriminant score = innovation propensity index 0.446 x market scope + 0.003 x number of client countries - 0.099 x competitive position + 0.084 x innovation intensity - 0.072 x relative innovation + 0.272 x innovation effect on competitive advanatage + 0.002 x innovation importance + 0.901 x export to developed countries + 0.362 x trigger factors group + 0.169 x innovation effect on increase differentiation
Conclusions Active/Passive firms – behavior profiles higher/lower propensity to innovate related to market and innovation variabels possible reinforcement mechanism Suggest alternative way of assessing innovation in services indirect - measures innovation as a latent variable combines several indicators - more robust Sistematic bias – more innovative firms are more conservative in their evaluation of their innovation activities
Future Developments Apply the discriminant score to a set of known firms Larger data sets Other sectors in services and manufacturing
Thank you for your attention