Organizational Learning Facilitators in the Canadian Public Sector

Abstract

Organizational learning (OL) is considered to be a central element in the renewal of Canada's federal public service. What factors facilitate OL in this sector? How can these factors be measured? This study aims to answer these questions by describing the development of an instrument designed to produce a valid measure of the organizational learning facilitators (OLFs) relevant to public sector organizations. The confirmatory analysis indicated a 6-factor solution with 5 first-order factors (“knowledge acquisition and transformation,” “learning support,” “earning culture,” “learning leadership, and “strategic management”) and one second-order factor (“learning environment”). Results indicate that the OLF measure is a significant predictor of organizational outcomes.

Keywords:

• organizational learning
• facilitators
• executives
• public sector
• Canada
• confirmatory factor analysis

Notes

¹The qualitative approach is phenomenological in the sense that emphasis must be placed on the actors' experience in order to prevent researchers from imposing their own a priori scheme of reference in the emergence of information (Glaser & Strauss, 1967).

²The EX occupational classification represents to the highest levels of management in the Canadian public service.

³The following questions were asked:

a.	Describe in your own words what OL consists of?
b.	Could you illustrate this point with a specific example?
c.	What are the prerequisites for OL in your organization?
d.	In your opinion, what are the main facilitators of OL in your organization?
e.	From an individual and organizational standpoint, what are the factors confirming that OL has taken place in your organization?

⁴NVivo is a qualitative data analysis (QDA) computer software package produced by QSR International. It has been designed for qualitative researchers working with very rich text-based and/or multimedia information, where deep levels of analysis on small or large volumes of data are required.

⁵The experts familiar with the OL concept consisted of 5 professors in an OL research project, one consultant involved in introducing OL into a public organization, and 3 senior managers familiar with the public service.

⁶CVR = (n – N/2)/N/2 where n = the number of panellists who rated the item as 1 or 2 and N the total number of respondents. This procedure uses a majority vote to validate a given item's content. The CVR coefficient is interpreted as a correlation index.

⁷EFA is a statistical technique to achieve data reduction through a series of statistical parameters in order to achieve a sound representation of a given latent construct either through a simple (undimensionality) or more complex (multidimensionality) configuration of its constitutive items.

⁸For practical purposes, very small multivariate kurtosis values (e.g., less than 1.00) are considered negligible, while values ranging from 1 to 10 often indicate moderate non-normality. Values that exceed 10 indicate severe non-normality.

⁹Although this study does not specifically validate this scale of measurement, it is important to note that we have taken care of prior assessment parameters to measure of this instrument's psychometric properties. As such, the results observed following the confirmatory factor analysis indicate a good fit to the data (GFI = 0.95, CFI = 0.95, TLI = 0.94; RMSEA = 0.06). For these measurement scales, coefficients of internal consistency (Cronbach's alphas) obtained are 0.83 (individual learning), 0.80 (group learning), 0.85 (organizational impacts).

¹⁰ Bapuji and Crossman (2004) emphasize the need for more qualitative studies on OL in order to produce theories based on the reality of the environments involved. Only two empirical studies out of the 55 listed by these authors combined both qualitative and quantitative approaches in researching the OL phenomenon.