Correction to: Do professional staff in universities really challenge academic norms? A perspective from the Netherlands (Higher Education, (2024), 88, 6, (2165-2186), 10.1007/s10734-024-01209-x)

Several corrections need to be made in this article. They are as follows: 1. In Data collection section, paragraph 2, “N = 124” should be “N=118” 2. In Data collection section, paragraph 5, “four” should be “six”, “two” to “six”, “role and years of experience, respectively” to “(years of experience (N=1), strength of relationships with government (N = 4), and role (N = 1), respectively”, “the analyses were conducted based on a final sample of N = 124” should be “the hypotheses were tested based on a final sample of N = 118” 3. In Variable section, paragraph 1, “ten” should be “twelve”, “the initial solution indicated five factors, three of them having an eigenvalue greater than one (KMO = 0.749; Bartlett’s test of sphericity indicated that the correlation matrix underlying the factor analysis is significantly different from zero (χ2 (153) = 735.75, p < 0.001)). After rotation, a simple structure based on those three factors was obtained. Even though the items within the scales differed from those identified by Shields and Watermeyer (2020), two of the scales clearly indicated an academic professional and a market logic. The third scale, matching with the theoreticalconstruct of the state logic, was identified but contained only two items, thus not suited for measuring the construct (MacCallum et al., 1999)” to “the initial solution indicated three factors, all with an Eigenvalue greater than one (KMO = 0.809; Bartlett's test of sphericity indicated that the correlation matrix underlying the factor analysis is significantly different from zero (χ²(120) = 660.82, p < 0.001)). After rotation, a simple structure based on those three factors was obtained. Even though some of the items within the scales differed from those identified by Shields and Watermeyer (2020), two of the scales clearly indicated an academic professional and a market logic. The third scale included two items with negative loadings larger than −0.4 that seem to indicate a community logic, which is not included as a variable in this study (Thornton et al., 2012)” 4. In Variable section, paragraph 2, “Cronbach’s alpha of 0.75 and a weaker, but still moderate and thereby acceptable, reliability of 0.62 for the academic professional logic” to “Cronbach’s alpha of 0.79 and a weaker, but still moderate and thereby acceptable, reliability of 0.62 for the academic professional logic when only including items that uniquely load on this scale” 5. New Tables 1, 2, 3 and 4 6. In Analysis section, paragraph 1, “estimation. Missing data was deleted listwise, which affected one out of 124 cases on the independent variable private sector experience when testing hypotheses 1 and 3.” to “based on standardized scores.” 7. In Results section, paragraph 1, “respondents (m = 55, SD = 20.4). Their academic professional logic is high, on average, and the variation between respondents is smaller (m = 85.6, SD = 11.4). Business developers have a significantly higher mean on the market logic than the other PS roles (ANOVA, η²= 0.12, F(3,121) = 4.6, p = 0.004).” should be “respondents (m = 47.68 SD = 22.43). Their academic professional logic is high, on average, and the variation between respondents is smaller (m = 86.18, SD = 11.11). Business developers have a significantly higher mean on the market logic than the other PS roles (ANOVA, η²= 0.106 F(3,119 = 4.7, p = 0.004).” 8. In Results section, paragraph 2, “sector experience (ρ = 0.44, p < 0.01) and relationships with private companies (r = 0.37, p < 0.01) and the dependent variable market logics. There is also a weak positive relationship between the two independent variables (ρ = 0.33, p < 0.01).” should be “sector experience (ρ = 0.38, p < 0.01) and relationships with private companies (r = 0.48, p < 0.01) and the dependent variable market logics. There is also a weak positive relationship between the two independent variables (ρ = 0.36, p < 0.01).” 9. In Results section, paragraphs 4, “The first model, including only the dependent variable and control variables, explains 22% of variance in the dependent variable, which is significantly different from a model with zero predictors (R2 = 0.22, F(12,111) = 2.61, p = 0.004). Except for the management position, no control variables show a significant effect on the market logic of PS. Holding a management position has a significant positive effect on the dependent variable (b = 5.20, t = 2.59, p = 0.011, CI98.8 = (0.10, 10.29)). Adding the independent variable private sector experience does significantly improve the model (R2-change = 0.07, F-change(1,109) = 9.88, p = 0.002). There is a medium positive effect of private sector experience (f2 = 0.30) indicating that PS with private sector experience score higher on market logic. This effect is significant at the corrected alpha level of 1.25% (b = 6.09, t = 3.14, p = 0.002, CI98.8 = (1.17, 11.01)). Thus, hypothesis 1 is supported. In the second model, no control variables have a significant effect on the dependent variable at the pre-defined 1.25% level.” should be “The first model, including only the dependent variable and control variables, explains 28% of variance in the dependent variable, which is significantly different from a model with zero predictors (R²= 0.28, F(12,105) = 3.44, p = < 0.001). No control variables show a significant effect on the market logic of PS. Adding the independent variable private sector experience does significantly improve the model (R²-change = 0.06, F-change(1,104) = 9.18, p = 0.003). There is a small positive effect of private sector experience (f²= 0.09) indicating that PS with private sector experience score higher on market logic. This effect is significant at the corrected alpha-level of 1.25%. (b = 6.31, t = 3.03, p = 0.003, CI98.8 = (1.02, 11.60)). Thus, hypothesis 1 is supported. In the second model, no control variables show a significant effect on the market logic of PS. The second hypothesis stated that there is a positive linear relationship between PS’s external relationships with private companies and PS’s market logic. Adding the independent variable strength of relationships with private sector companies to the first model in model 5 does significantly improve the model at the corrected alpha-level of 1.25% (R²-change = 0.06, F-change(1,104) = 9.02, p = 0.03). There is a small positive (f2 = 0.09) effect on the dependent variable, indicating that the stronger PS’s relationships with private companies is, the higher their market logic. This effect is significant at the corrected alpha-level of 1.25%. (b = 8.12, t = 3.00, p = 0.003, CI98.8 = (1.25, 14.99)). Hence, hypothesis 2 is also supported. In the fifth model, no control variables show a significant effect on the market logic of PS at the pre-defined 1.25% level.” 10. In Results section, paragraphs 5, combined with paragraph 4 11. In Results section, paragraphs 6, “The independent variable strength of internal relationships was introduced in the third model to measure its direct effect. This did not result in a significant increase in the explained variance in the dependent variable (R2-change = 0.008, F-change(1,108) = 1.22, p = 0.272). In the fourth model, an interaction term private sector experience*strength of internal relationships was added to model 3 to test hypothesis 3 that stated that the effect of private sector experience on market logic depends on the strength of internal relationships of PS. The interaction term did not increase the explained variance in the dependent variable (R2-change = 0.000, F-change(1,107) = 0.07, p = 0.794). Based on this negligible (f2 = 0.00) and not significant negative effect of the interaction term (b = − 0.48, t = − 0.262, p = 0.794, CI98.8 = (− 4.89, 3.80)), the null hypothesis cannot be rejected. Therefore, the third hypothesis is not accepted.” should be “The independent variable strength of internal relationships was introduced in the third model to measure its direct effect. This did not result in a significant increase in the explained variance in the dependent variable (R²-change = 0.01, F-change(1,103) = 1.538, p = 0.218). In the fourth model, an interaction term private sector experience*strength of internal relationships was added to model 3 to test hypothesis 3 that stated that the effect of private sector experience on market logic depends on the strength of internal relationships of PS. The interaction term did not increase the explained variance in the dependent variable (R²-change = 0.002, F-change(1,102) = 0.358, p = 0.551). Based on this small (f2 = 0.02) and not significant negative effect of the interaction term (b = −1.08, t = −0.599, p = 0.551, CI98.8 = (−5.67, 3.51)), the null-hypothesis cannot be rejected. Therefore, the third hypothesis is not accepted.” 12. In Results section, paragraphs 7, “Model 6 introduced the independent variable strength of internal relationships to model 5 to measure its direct effect on the dependent variable, which did not result in a significant increase in the explained variance in the dependent variable (R2-change = 0.002, F-change(1,109) = 0.34, p = 0.560). Lastly, the interaction term strength of relationships with private sector companies*strength of internal relations in model 7 was added to model 6 to test whether the effect of relationships with private sector companies on market logic is dependent on the strength of internal relationships of PS as hypothesis 4 stated. In the resulting seventh model, the interaction term did not significantly increase the explained variance in the dependent variable (R2-change = 0.003, F-change(1,108) = 0.48, p = 0.488). Based on this medium (f2 = 0.06) positive, but not significant effect of the interaction term (b = − 1.36, t = − 0.70, p = 0.488, CI98.3 = (− 6.34, 3.61)), the null hypothesis cannot be rejected. Thus, the fourth hypothesis is also not accepted.” should be “Model 6 introduced the independent variable strength of internal relationships to model 5 to measure its direct effect on the dependent variable, which did not result in a significant increase in the explained variance in the dependent variable (R²-change = 0.003, F-change(1,103) = 0.54, p = 0.464). Lastly, the interaction term strength of relationships with private sector companies*strength of internal relations in model 7 was added to model 6 to test whether the effect of relationships with private sector companies on market logic is dependent on the strength of internal relationships of PS as hypothesis 4 stated. In the resulting seventh model, the interaction term did not significantly increase the explained variance in the dependent variable (R²-change = 0.006, F-change(1,102) = 0.99, p = 0.322). Based on this medium (f2 = 0.02) positive, but not significant effect of the interaction term (b = −2.01, t = −0.99, p = 0.322, CI98.3 = (−7.14, 3.13)), the null-hypothesis cannot be rejected. Thus, the fourth hypothesis is also not accepted.” 13. In Interpretation of results section, paragraph 2, “As the effect size for the influence of relationships with private sector companies is small,” should be “As the effect sizes for the influence of private sector experience and relationships with private sector companies are small,” 14. Delete reference MacCallum et al.