How to draw a Mohr circle for the load

Development of a questionnaire to record key features of Work 4.0


Factor structure

To review research question 1, the factor structure in sample 1 was examined as part of two exploratory factor analyzes (EFA, principal axis analysis with Promax rotation). The number of factors to be extracted was determined via parallel analysis and the minimum average partial test (MAP test), since both procedures have proven to be superior to other techniques (Fabrigar, Wegener, MacCallum & Strahan, 1999; Ruscio & Roche, 2012). The sample proved to be suitable for carrying out an EFA (KMO = .80, Bartlett test: χ² = 8774.76, p <.001). In the first EFA indexed MAP test and parallel analysis, the extraction of nine factors matched. Six of the factors could be interpreted in terms of content. Five of the factors also had a homogeneous structure, whereas the sixth factor consisted of four items with related content with a high load and otherwise of secondary loads of other factors. The three other factors could be interpreted as method artifacts, as only two to three items were loaded on each of them. On the basis of the first EFA, items with high secondary loads (> .30) and low factor loads (<.30) or communalities (<.20) were excluded (cf. Bühner, 2011). In order to ensure adequate internal consistency of the scales and at the same time to guarantee a short response time, five items were selected for each construct on the basis of statistical parameters and the content-related considerations of the first author and the second author. The resulting 25 items were examined with a second EFA. MAP test and parallel analysis identified five factors to be extracted. The factor loadings of the final item selection were between .35 ≤ Hi ≤ .97 and can be clearly assigned to the respective factor. A total of five items could be assigned to each factor. Only item 3 of the flexibilization factor “I answer work e-mails outside of the office (e.g. at home, on the train)” showed a significant sideload of Hi = .30 on the delimitation factor.

To replicate the results, the EFA was repeated with the data from sample 2. The sample proved to be suitable for carrying out the EFA (KMO = .78, Bartlett test: χ² = 1743.26, p <.001). For this sample, too, the MAP test and parallel analysis indicated the extraction of five factors. The charge pattern was comparable to that from the first subsample. For item 3 of the flexibilisation factor “I also answer work e-mails outside of the office (e.g. at home, on the train)”, there was another sideline of Hi = .33 on the delimitation factor. In addition, item 1 “I can organize my working hours flexibly” showed only a small load on the associated flexibilization factor. In order to ensure a sufficient breadth of the content of the factor, however, the item was not excluded. The results of both factor analyzes are shown in Table 2.

Item No.M.SDFactor 1Factor 2Factor 3Factor 4Factor 5H2
1 Digi3.98/4.071.32/1.32.73/.67.57/.52
2 digi3.79/3.721.49/1.54.58/.44.40/.31
3 digi 3.72/3.721.39/1.51.91/.81.77/.66
4 digi3.12/3.081.38/1.47.64/.65.41/.43
5 digi3.80/3.861.43/1.45.80/.78.62/.60
6 flex3.25/3.221.42/1.35.35/.22.34/.15
7 flex1.94/2.071.35/1.42.69/.77.54/.59
8 flex2.85/2.901.59/1.63.56/.48.30/.33.51/.40
9 Flex2.38/2.511.37/1.40.79/.83.58/.66
10 flex3.23/3.211.54/1.47.77/.59.51/.35
11 deburring2.52/2.541.24/1.30.85/.82.64/.64
12 deburring3.04/3.161.28/1.38.66/.64.40/.46
13 deburring2.77/3.011.42/1.45.51/.58.46/.36
14 deburring2.51/2.411.35/1.39.57/.55.31/.31
15 deburring2.73/2.921.30/1.37