DISCUSSION THE TEST REVEALED TWO HOMOGENOUS SUBSETS, WHICH SUGGESTE...

6. Discussion

The test revealed two homogenous

subsets, which suggested that attitude to

The purpose of the study was to test

responses with the influence of a binary

components of the theoretical model

question structure (Subset 1, Table 3)

proposed by Bircham (2003) and to

differed to those with the influence of an

address the question of whether or not

open-ended or directed question structure

question structure is of importance in the

(Subset 2, Table 3). The difference

knowledge sharing process. The results of

between open-ended and directed

the study support the notion that question

questions structures for the factor attitude

structure does matter; questions of a

was small and not significant (p<0.01).

binary structure had a lower attitude

measure than those questions for either

The Likert scale instrument labels ranged

an open-ended or directed structure.

from strongly disagree = 1 through neutral

These findings complement the literature

= 4 to strongly agree = 7 and the

on knowledge sharing by answering the

calculated means for the attitude measure

call for further investigation into

increased when question structure

questioning (Cooper, 2003) as well as

complexity increased (binary to open-

focusing on factors that may influence the

ended and directed). This suggests that

recipient.

the recipients were more favourably

disposed towards the knowledge they

There are some potential limitations to the

received when questions of a complex

study. First, the question designer was not

structure were used (Figure 1). This

the same individual as the recipient of the

finding supports hypotheses H

1a

and H

1c.

responses. If the two were the same

However, there was no significant

individual then potentially the findings of

difference in attitude between questions of

this study may not hold. For instance, if

an open-ended or directed structure, even

the person who designs and asks the

though directed questions had a slightly

question is also the recipient of the

higher attitude measure. Therefore,

responses, then in many circumstances it

hypothesis H

1b

is not supported.

would not be unreasonable to assume that

they already possess substantial

knowledge associated with the domain of

the question. The question response

structure preferred in this circumstance

could be of the closed type, rather than

open-ended or directed. However, in many

organisations, if not the majority, the

person required to make the decision,

based on the knowledge received, is not

Figure 1: Mean Plots from Tukey HSD for

the same individual as the question

Attitude

designer (e.g. a finance director may make

the decision and an auditor may design

received. While it is not possible to

definitively conclude from the results of

the questions).

this study that this increase reflects more

favourable attitude in the recipient towards

Second, while this research has

established that question structure does

the knowledge received, neither can such

a conclusion be confidently dismissed.

influence attitude, there is a potential

The findings of this study strongly indicate

limitation in the attitude measure. The

purpose of Fishbein and Ajzen’s (1975)

that a recipient’s attitude towards

knowledge received varies with the

model was to predict behaviour given

structure of the questions used to elicit

attitude and intentions. While the 5-item

attitude measure has been successfully

knowledge from a source. Therefore,

used in various studies (many external to

understanding the influence of question

structure in knowledge sharing is

the field of knowledge management) future

potentially of major significance to

investigation into a recipient’s satisfaction

business and government.

with the knowledge may result in the

development of a more vigorous measure.

For example, a recent development and

Acknowledgements

validation of a measure for website user

This research was funded in part by The

satisfaction (Muylle, Moenaert, &

Foundation of Research Science and

Despontin, 2004) expanded the definition

Technology, New Zealand.

of satisfaction to include identifying

underlying dimensions of the construct,

References

inclusive of comprehensibility,

comprehensiveness, accuracy, relevance

Alvai, M., & Leidner, D. (2001). Knowledge

and format. Future research could expand

management and knowledge

the definition of satisfaction with

management systems:

knowledge received to include such

Conceptual foundations and

underlying aspects.

research issues. Management

Information Systems, 25(1), 107-

Finally, the constructs attitude and