Results | Trust in Knowledge Management and Systems in Organizations

< Day Day Up >

Descriptive Statistics

Frequencies and percentages were calculated to summarize the nine contingency variables, including the situational and demographic characteristics of the respondents (Table 3). While 115 (53.7%) of the 214 responding users were from the public sector, 99 (46.3%) were from the private sector. As to the manufacturing sector, 78 (36.4%) of the respondents were in textiles, 70 (32.7%) were in chemicals, and 66 (30.8%) were in the electronics and engineering industries. Ninety three (43.5%) of the respondents were top level manages, 105 (49.1 %) were middle level managers, and the remaining 16 (1.5%) were lower level managers.

Table 3: The Contingency Variables (user's situational and demographic characteristics).
Respondents' Characteristics	The Overall Sample
Respondents' Characteristics	Freq.	%
Ownership Type:
Public	115	53.7
Private	99	46.3
Total	214	100
Industry Type:
Textiles	87	39.0
Chemical	70	31.4
Electronics & Engineering	66	29.6
Total	223	100
Managerial Level:
Top management	93	43.4
Middle management	105	49.1
Lower management	16	7.5
Total	214	100
Functional Area:
Finance	63	29.4
Production	57	26.6
Sales & marketing	39	18.2
Human resources	33	15.5
Logistics	22	10.3
Total	214	100
Educational Level:
Graduate	8	3.7
Undergraduate	166	77.6
High school	40	18.7
Total	214	100
Educational Background:
Business	138	64.5
Engineering	44	20.6
Science	17	7.9
Law	4	1.9
Arts	5	2.3
Others	6	2.8
Total	214	100
Tenure in the Organization:
1–5 years	36	16.8
6–10	44	20.6
11–15	25	11.7
16–20	17	7.9
21–25	21	9.8
26–30	22	10.3
31–35	44	20.6
36–40	5	2.3
Total	214	100
Computer Training:
Yes	115	53.7
No	99	46.3
Total	214	100
System Age in User Departments:
1–3 years	55	25.7
4–7	99	46.3
8–11	40	18.7
12–15	12	5.6
16–18	3	1.4
19–22	5	2.3
Total	214	100

With regard to the respondents' functional areas, 63 (29.4%) were from finance, 57 (26.6%) were from production, 39 (18.2%) were from sales and marketing, 22 (10.3%) were from logistics, and 33 (15.4%) were from human resources. Eight (3.7%) reported some graduate education, 166 (77.6 %) held college degrees, and 40 (18.7 %) had a high school or equivalent degrees. One hundred thirty eight (64.5%) of the respondents had a business background, 44 (20.6%) had an engineering background, 17 (7.9%) had a science background, and the remaining 15 (7%) were from other educational backgrounds. More than 62% of the users had tenures of eleven years or more in the companies they were working with at the time of the study, and only 53.7% of them had formal computing training. Also, more than 90% of the systems investigated were in use for eleven years or less.

Table 4 represents a summary of descriptive statistics for the variables in the study. The respondents (managers) generally agreed upon the existence of a relatively high level of top management support for IS, and on the existence of an average to high level of both user involvement and systems maturity in their companies. Also, the respondents indicated a rather high level of information satisfaction and a high level of system use in improving decision making as the two measures of systems effectiveness.

Table 4: Descriptive Statistics for the Research Variables.
Research Variables	Mean	Standard Deviation	Minimum	Maximum
Top management support	4.23	.62	1.67	5
IS maturity	3.62	.74	1.70	5
User involvement	3.60	.82	1.00	5
System use in improving decision making	4.04	.66	1.00	5
User information satisfaction	4.01	.60	1.31	5

Hypotheses Testing

The research hypotheses were tested using Pearson's correlation and simple regression analysis. Table 5 shows a summary of the hypotheses testing results. As to H1, the analysis revealed a significantly positive correlation between top management support and user information satisfaction (r=.33, P<.001). This finding supports the acceptance of H1, indicating that the greater the top management support, the greater the user information satisfaction. Top management support explained 11% of the variance in user information satisfaction(F = 27.26, P<.001).

Table 5: Results of Hypotheses Testing Using Pearson's Correlation and Simple Regression Analysis.
*Hypotheses*	*Independent*	*Dependent*	R	R²	F
	Variable	Variable
H1	Top management support	User information satisfaction	.33***	.11***	27.26
H2	Top management support	System use in improving decision making	.35***	.12***	30.45

H3	User involvement	User information satisfaction	.37***	.13***	34.44
H4	User involvement	System use in improving decision making	.34***	.12***	29.53

H5	IS maturity	User information satisfaction	.58***	.33***	107.57
H6	IS maturity	System use in improving decision making	.44***	.19***	51.07

With regard to H2, the analysis revealed a significant and positive correlation between top management support and system use in improving decision making (r = .35, P <.001). This finding supports the acceptance of H2, which means the greater the top management support, the greater the improvement in decision making through the use of information systems. Top management support explained 12% of the variance in system use in improving decision making (F= 30.45 and P <.00.1).

The testing of H3 shows a significant and positive correlation between user involvement and user information satisfaction (r =.37 and P <.001). This finding supports the acceptance of H3, which means the greater the user involvement in the definition, design, and implementation of a system, the greater the user information satisfaction. User involvement explained 13% of the variance in user information satisfaction (F = 34.44 and P <.00.1).

Testing of H4 revealed a significant and positive correlation between user involvement in the definition, design, and implementation of a system, and system use in improving decision making (r = .34, P <.001). This finding supports the acceptance of H4. The greater the users' involvement in the definition, design, and implementation of a system, the greater the use of the system in improving decision making. User involvement explained 12% of the variance in systems use in improving decision making (F = 29.53, P <.001).

As for IS maturity, the testing of H5 indicated a significant and positive correlation between IS maturity and user information satisfaction (r=.58, P <.001). This finding supports the acceptance of H5, which means the greater the maturity of IS, the greater the users' information satisfaction. IS maturity explained 33% of the variance in user information satisfaction (F = 107.57, P <.001).

The results of H6 testing show a significant and positive correlation between IS maturity and system use in improving decision making (r = .44, P <.001). This finding supports the acceptance of H6, which means the greater the maturity of IS, the greater the use of systems in improving decision making. Also, IS maturity explained 19% of the variance in systems use in improving decision making (F = 51.05, P <.001).

To further examine the importance of the three independent variables as predictors of system effectiveness, a stepwise multiple regression analysis was performed. The stepwise multiple regress analysis was used because of the lack of a well-defined predictive model of systems effectiveness. The results of the final step of the analysis are shown in Table 6.

Table 6: Stepwise Multiple Regression Analysis: The Predictors of System Effectiveness.
*Dependent Variables*	*User Information Satisfaction*		*System Use in Improving Decision Making*
Predictor Variables	R	∆r²	R	∆r²
IS maturity	.58	.33 ^[***]	.44	.19^[***]
User involvement	.59	.02^[***]	.47	.03^[***]
r²		.35^[***]		.22^[***]
n = 214
^[**]P < .001*

From Table 6, the variables in the regression equation (IS maturity and user involvement) cumulatively explained .35 and .22 of the variance in user information satisfaction and system use in improving decision making, respectively. The incremental contributions (r²) of IS maturity in explaining the variance in user information satisfaction and systemuse in improving decision making are .33 and .19, respectively. However, only .02 of the variance in user information satisfaction and .03 of the variance in system use in improving decision making are explained respectively by user involvement.

The research model in Figure 1 implies that the effect of top management support, user involvement, and IS maturity on system effectiveness could be contingent upon factors such as system age in user department, user's functional area, user's tenure in the organization, user's organizational position, user's educational level, user's educational background, user's computer training, type of industry, and type of ownership.

To examine the impact of these contingency variables, an analysis of variance and covariance using the independent and contingency variables was performed (Table 7). The direct effects of the independent variables (top management support, user involvement, and IS maturity) on user information satisfaction and system use in improving decision making were significant (F = 11.369, P < .01 and F = 6.282, P < .01, respectively). However, the effect of the contingency variables was significant only on user information satisfaction. After taking these contingency variables into account, the explained variance in user information satisfaction increased by 5.751 (F = 3.002, P < .05).

Table 7: Analysis of Variance of Independent Variables and Contingency Variables on Ssystems Effectiveness.
Source of variation	Dependent variables	Sums of squares	DF	Means squares	F
Main effects (top management support, user involvement, and IS maturity)	User information satisfaction	29.046	12	2.420	11.369***
Covariates (contingency variables):		5.751	9	.639	3.002.**
User's functional area		.050	1	.050	.233
Ownership type		1.800	1	1.800	8.456**
User's organizational position		.209	1	.209	.981
User's tenure in the organization		2.120	1	2.120	9.956**
System age in user department		.108	1	.108	.505
User's educational level		.472	1	.472	2.216
User's educational background		.282	1	.282	1.323
User's formal computer training		.034	1	.034	.158
Industry type		.118	1	.118	.553
Main effects (top management support, user involvement, and IS maturity)	System use in improving decision making	25.146	12	2.096	6.282***
Covariates (contingency variables):		2.962	9	.329	.987
User's functional areas		.105	1	.105	.316
Ownership type		.297	1	.297	.889
User's organizational position		.359	1	.359	1.075
User's tenure in the organization		.022	1	.022	.066
System age in user department		.173	1	.173	.517
User's educational level		.657	1	.657	1.969
User's educational background		.579	1	.579	1.736
User's formal computer training		.194	1	.194	.582
Industry type		.184	1	.184	.552

User's tenure in the organization and ownership type (public vs. private) were the only two contingency variables that affected the explained variance in user information satisfaction. User's tenure in the organization explained the variance in user information satisfaction by 2.120 (F = 9.956, p < .05), and ownership type explained the variance in user information satisfaction by 1.800 (F = 8.456, p < .05). The effects of the other contingency variables on user information satisfaction were insignificant.

< Day Day Up >