Development of Intercultural Competence After Nine Months of Study
Does the level of intercultural competence of ﬁrst year master students increase whilst on campus during the ﬁrst nine months of study at University? The results of the IDI indicate that the development orientation of the total sample lies in early Minimization at the pre-test, as well as at the post test assessment. The benchmark group and the unknown group score at the cusp of Minimization at the preand the post-test. The mean score of the new entrants is in Minimization. The mean IDI scores of the benchmark, the new entrants and the unknown groups for the pre-test and the post-test are given in Table 2.
The data in Fig. 2 inform the mean scores of each of the three groups and the total sample slightly decreased at the post-test. However, the development orientation is unchanged and remains in early Minimization. Regardless whether the respondents progress from an undergraduate program or are new to the university, t-tests for paired sample means could not conﬁrm signiﬁcant differences within each of the groups between the pre-test and the post-test. Furthermore, no signiﬁcant differences in IDI scores between the groups can be observed after 10 months.
The IDI scores indicate the developmental orientation of the respondents. The IDI orientations of the benchmark and the group of new entrants compared to the population distribution for the pre-test are given in Fig. 3. The two test groups  appear to deviate from the population distribution. In the benchmark group, Denial and Polarization seem over-represented, and in the group of the new entrants Polarization seems over-represented. The global worldviews Acceptance and Adaptation are under-represented in the pre-test.
The post-test distribution seems to suggest the distributions of the benchmark group and the group of new entrants across the IDI Orientations have become more similar, however still deviate from the population distribution. Although there are some exceptions, in Fig. 4 a general tendency towards Polarization can be observed. A series of Kolmogorov-Smirnov test for small samples has been performed, assessing whether the observed  frequencies of the benchmark group, the group of new entrants and the total sample match the population distribution. The tests conﬁrm that Polarization is over-represented and Acceptance and Adaptation under-represented for both groups at the pre-test, as well as at the post-test. A tendency towards polarization could not be conﬁrmed  for the test groups separately. However, at the level of the total sample, a signiﬁcant difference  has
Fig. 2 The mean IDI raw DO scores of the benchmark group, the new entrants group and the Unknown compared to sample and the population mean
Fig. 3 The relative distribution of the pre-test IDI scores across the IDI orientations
Fig. 4 The relative distribution of the post-test IDI scores across the IDI orientations
been found indicating that more respondents with a pre-test score in Minimization regressed to Polarization in the post-test than expected  if change had been random.
In the benchmark group, change in development orientation and the direction of that change occur randomly. In the group of new entrants, signiﬁcantly fewer participants changed in IDI orientation then if change had been random ; the direction of change however is random. For the total sample, the number of respondents who changed in orientation is lower than expected then if change had been random ; in case change takes place, the direction is random.
The range of available IDI Orientations within the sample at the pre-test and the post-test is relatively narrow primarily in the Mono-cultural and Minimization areas of the IDI scale.
The results so far indicate that the level of intercultural competence in terms of IDI scores does not increase in the 9 month study period, regardless whether respondents are new to the university. However, a tendency in development orientation towards Polarization has been conﬁrmed. To better understand this tendency, an analysis for each of the pre-test IDI Orientation has been performed. The T-test for correlated samples conﬁrms that the group of respondents (n = 10) with a pre-test development orientation of Denial progressed in the IDI scores at the post-test. Six of these respondents progressed towards Polarization; one towards Minimization. No signiﬁcant change in IDI scores has been identiﬁed for the group of respondents with a pre-test development orientation in Polarization; although six respondents actually regressed into Denial; and three respondents progressed towards Minimization. The group of respondents with a pre-test development orientation in Minimization regressed to Polarization. This primarily can be attributed to the subgroup of respondents in early Minimization. No signiﬁcant change in IDI scores has been identiﬁed for the group of respondents who scored in Acceptance at the pre-test.
-  BM group: T = 0.18; df = 30; p = 0.86. NE group: T = 1.35; df = 53; p = 0.18. Unknown group: T = 1.52; df = 22; p = 0.14
-  ANOVA single Factor Analysis; T1: F = 0.78, p = 0.46; T2: F = 0.45, p = 0.64
-  The group of respondents of which it is not known whether they are new to the university has been excluded from the interval analysis because of the n < 30 and because the t-tests concluded there were no signiﬁcant differences between the Unknown and the two test groups
-  A value of 0.5 has been added to each of the cells in the frequency table to avoid the empty cells. According to Agresti (1990), this enables the statistical analysis but does not influence the actual outcome, (p. 54). The reworked number of respondents is referred to as n′ and equals 33.5
-  McNemar's test for correlated samples failed to conﬁrm statistical signiﬁcance. In the benchmark group one case, in the group of new entrants three cases have been eliminated to enable this test. These cases were considered outliers
-  (p = 4.34E-03)
-  A random probability of change in development orientation refers to a 50 % chance that a respondent remains in a developmental orientation and a −50 % chance that a respondent increases or decreases in developmental orientation
-  The direction of change refers to a decrease or an increase in development orientation
-  Conﬁrmed by exact binominal calculations
-  (p = 0.04)
-  (p = 4.53E-03)
-  (p = 4.12E-05)
-  (p = 1.45E-02)
-  (p = 3.80E-02)