11.3 Cross-cultural comparison of Austrian, Czech and Polish Managers

11.3.1 Methodology and data collection

The methodology and data collection were dominated by a clear action orientation. No questionnaire was used and all the data were collected by administering a ˜problem set in the form of 30 decision-making situations. The 30 situations were selected and rewritten from descriptions of real decisions provided by hundreds of real managers and were validated with the assistance of trained managers (Vroom et al ., 1976). If eight out of 10 of those trained managers detected the same problem, the same case, sufficient validation was assumed. This test was first applied in English (Jago and Vroom, 1978) and was later repeated in a German version (B hnisch, 1991). For the Czech and Polish studies a translation of the 30 situations by native speakers was used. The semantic ˜corrections were not tested systematically. However in discussions with the Czech managers during their training programme, one of the authors in charge of the feedback session got the impression that the translation has a high degree of validity because the managers found the same problem attributes for each case. The reliability of the Polish version of the problem set was tested for half the cases (15). Based on a sample of 121 Polish managers this test produced adequate results (Maczynski et al ., 1997).

The problem set was administered to managers who, at the time of data collection, were unfamiliar with the Vroom “Yetton model. In addition to the 30 decision-making situations they were presented with five strategies and asked to select one for each situation. An average time of two hours was needed to study the situations and make their decisions.

The results of the decision-making process mirrored intentions. The validation studies conducted by Jago and Vroom (1978) for the US and replicated by B hnisch et al . (1988) for Austria concluded that the intentions given in response to the problem set were equivalent to the actual behaviour of the managers concerned .

The Czech and Austrian data were collected prior to the holding of leadership training programmes. In such programmes, the respondents did not work through the problem set to do a ˜favour to the researchers “ but were mainly concerned to improve their own leadership behaviour. All of the participants received feedback, in which their first reactions to the problem set were compared with a description of the model. Training was provided to help the participants to use the diagnostic questions and decision rules for leadership decisions in their own organizational environment.

The Polish data stemmed from two prior studies. The data were collected in 1988 from 146 managers (Maczynski et al., 1994) and in 1993 “94 from 253 managers (Jago et al ., 1996). In this case the data collection was not completed within the framework of a training programme. The results received some underpinning by the study by Szabo et al . (1997), who used the value-oriented framework of the GLOBE project for 278 Polish and 169 Austrian middle managers.

The data collection began in 1984 in Austria and in 1991, in the Czech Republic, the most recent data being collected in Prague in spring 2002. The total numbers were standardized, based on a matching process. Matching was performed on organizational and demographic variables (provided by the respondents) that are known to affect leadership style: gender, hierarchical level, managerial function, age, organization type, number of subordinates , and tenure with the company.

11.3.2 Results

11.3.2.1 Participation

The most straightforward problem-set statistics are the frequency with which managers choose each of the five strategies. Table 11.1 shows the means for Austria, the Czech Republic and Poland. The comparison confirms the finding that Austrian managers are least inclined to employ autocratic strategies (AI and AII) and most frequently use group processes (CII and GII) for decision making. Polish and Czech managers do not differ significantly from each other in the use of autocratic strategies and consultative group processes (CII); managers in the Czech Republic are only different from their counterparts in Poland in that they use the GII strategy less frequently. As far as the CI strategy is concerned no differences can be found, and the use of this strategy in all three countries has the second lowest frequency. CI is the only strategy in which managers in all three countries showed no significant difference.

Table 11.1: Participation

	A (146)	CZ (146)	PL (146)	F-value	A vs CZ	Avs PL	CZ vs PL
Use of strategies (per cent) AI (autocratic) AII (autocratic) CI (Consultative) CII (Consultative) GII (group decision)	17.50 14.90 15.40 30.90 21.20	25.80 19.60 17.00 23.30 14.30	25.70 18.10 17.00 22.40 16.80	24.69** 10.65** 7.72 27.8** 18.24**	** ** “ ** **	** ** “ ** **	“ “ “ “ **
Mean level of participation	5.52	4.34	4.50	48.67**	**	**	“
Standard deviation	3.57	3.69	3.75	5.96**	**	**	“
*p <  0.05. **p < 0.01

Based on the participation score of the five leadership and their corresponding participation score strategies a mean level of participation (MLP) can be computed. In our study it is not computed at the individual level (which reveals a personality factor), but rather as an average at the national level. The standard deviation (SD) around the average is also computed at the national level. The SD indicates flexibility: the higher the score “ the maximum on the participation scale being between AI (0) and GII (10) “ the higher the variance of strategies.

The MLP scores and standard deviations are shown at the bottom of Table 11.1. As can be seen, the Czech Republic and Poland are united in having significantly lower MLP. It is interesting to note that the standard deviation for these countries is higher than for Austria. This confirms the assumption (based on the GLOBE data) that Polish and Czech managers possess a high degree of flexibility as a whole, which can be interpreted “ as already mentioned “ as a sign of readiness for the process of change.

Table 11.2: Agreement with the Vroom “Yetton model

	A (146)	CZ (146)	PL (146)	F-value	A vs CZ	Avs PL	CZ vs PL
Percentage conforming to:
Feasible set	73.6	65.7	64.1	68.08**	**	**	“
Model A choice	37.9	36.0	32.2	12.50**	“	**	**
Model B choice	30.8	20.3	21.3	51.84**	**	**	“
Rule violations (per cent)
Rule 1 (Leader info .)	8.6	13.0	17.4	22.55**	**	**	**
Rule 2 (goal congruence )	12.0	9.4	11.7	2.16	“	“	“
Rule 3 (unstructured)	33.9	48.4	50.8	24.36**	**	**	“
Rule 4 (acceptance)	15.1	28.9	32.1	49.17**	**	**	**
Rule 5 (conflict)	31.5	51.5	56.4	54.62**	**	**	**
Rule 6 (fairness)	23.9	56.9	48.3	18.47**	**	**	**
Rule 7 (accept priority)	58.4	73.3	75.7	23.01**	**	**	“
Quality rules (1 “3)	15.0	18.7	21.8	28.84**	**	**	**
Acceptance rules (4 “7)	28.9	45.4	42.7	77.80**	**	**	“
*p < 0.05.
**p < 0.01

11.3.2.2 Agreement with normative model

Table 11.2 shows the mean frequencies with which the Austrian, Czech and Polish responses fell within the feasible set across the 30 decision-making situations. Austrian managers outperform their counterparts in Poland and the Czech Republic, with a very high F-value. Austrian and Czech managers show the same degree of conformity to Model A (time efficiency) but there is a substantial difference between them in the case of Model B (subordinate development), where the scores of Polish and Czech managers lie close together.

Each time a respondent s choice lies outside the feasible set, that choice has violated one or more of the seven decision rules underlying the normative model. The rates of rule violation are reported in Table 11.2. These data isolate the sources of disagreement between managers and model behaviour. For six of the seven rules the Austrian respondents show a lower rate of violation than the Czech and Polish respondents; the Czech and Polish managers are congruent in five of the seven rules. Rule 2 “ which excludes the GII strategy in situations where quality is at stake and the subordinates do not share the organization s goals “ has one of the lowest frequencies of violations and is the only strategy for which there is no significant difference between the three countries.

As previously stated, rules 1 to 3 are designed to protect decision quality whereas rules 4 to 7 are designed to protect decision acceptance. The rates of quality rule violations (appropriately adjusted for the frequency of rule applicability) and acceptance rule violations are also included in Table 11.2.

The first conclusion to be drawn from the aggregation of the rule violations is that regardless of culture, departures from the model s prescriptions are more likely to be violations of acceptance rules than of quality rules. This is consistent with the finding in all studies completed within the framework of the Vroom “Yetton model (Vroom and Yetton, 1973; Vroom and Jago, 1988; Reber, et al ., 1993; Maczynski et al ., 1994; Reber et al ., 2000) and seems to have specific significance for the education of future managers. The deficits are significantly higher in the area of social skills than in professional skills. The latter refer to a narrow task orientation regarding the technical quality of a decision (for example: solving accounting problems, doing market research, scheduling the production process, etc.). Social skills have to do with social interaction, especially conflict resolutions .

Nonetheless there are significant differences among the three countries. Poland displays the highest rate of quality rule violations. Austria displays significantly lower rates of acceptance violation, for which Poland and the Czech Republic do not show significant differences.

11.3.2.3 Situational attributes “ main effects

Based on the seven diagnostic questions, the main effects are reported in Table 11.3. The main effects show behavioural differences that take place when an attribute is present versus absent and show the sensitivity to that particular attribute. A positive main effect indicates a tendency to be more participatory when the attribute is present (that is, when the answer to the diagnostic question is yes; a negative main effect indicates the reverse. The results present a relatively complicated picture. With regard to the quality requirement, Austrian and Czech managers are more participatory when the problem at hand includes a quality component and is, from the organization s perspective, not trivial, and they display greater autocracy in the case of organizationally trivial issues. Polish managers, however, display the opposite tendency. That is, they are more participatory in the case of trivial issues and more autocratic on substantial issues.

Table 11.3: Situational attributes “ main effects

	A (146)	CZ (146)	PL (146)	F-value	A vs CZ	Avs PL	CZ vs PL
Situational main effects:
Only requirement	0.49	0.86	“0.39	19.70**	“	**	**
Leader information	“0.32	“0.73	“1.10	11.89**	**	**	**
Problem structure	“2.00	“1.57	“0.86	14.08**	**	**	**
Acceptance requirement	0.79	0.85	0.32	7.09**	“	**	**
Prior prob. acceptance	“2.80	“2.33	“1.98	9.30**	**	**	“
Goal congruence	0.55	0.73	0.47	1.43	“	“	“
Subordinate conflict	0.16	“0.11	“0.40	7.31**	“	**	“
*p <  0.05.
**p < 0.01

In all three countries leaders become more participatory in situations where they do not have sufficient technical and/or professional information. This tendency is strongest among Polish managers ( “1.10), followed by Czechs ( “0.73) and Austrians ( “0.32). If the situation is unstructured, there is a tendency for autocratic behaviour to decrease in all three countries with significant differences between the three countries in the magnitude of the decrease: Austria ( “2.0), followed by Czech ( “1.56) and Poland ( “0.86). In situations where acceptance of subordinates is important, Czech (0.85) and Austrian managers (0.79) significantly tend to favour participative strategies to a higher degree than their Polish colleagues (0.32). In situations where leaders and subordinates are in conflict, Austrian managers ( “2.80) become significantly more participatory than managers in the Czech Republic ( “2.33) and Poland ( “1.98). If the conflict is between subordinates, managers from Poland ( “0.40) and the Czech Republic ( “0.11) become more autocratic, whereas Austrian managers (+0.16) become slightly more participatory.