Menu

Factor rotation Subsequent to the factor extraction

March 17, 2019 0 Comment

Factor rotation
Subsequent to the factor extraction, Varimax rotation (a type of Orthogonal rotation) was executed. It minimised the cross-products loadings. This minimisation yielded the simplified the factors which improved the interpretability of the retained components. The results, obtained after rotation, exhibited that the 33 variables were loaded on eight factors (Table 6.23).

Rotated Component Matrix Component Matrix
Com.
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8
Cont2 .714 .026 .039 .027 .089 .061 .044 .070 .493 .143 -.061 -.047 -.304 -.104 .382 .115 .531
Over1 .676 .143 .007 .179 -.020 .161 .043 .020 .549 .205 -.091 -.166 -.269 .017 .235 .177 .538
Cont1 .674 .079 .179 .150 .094 -.012 .018 .018 .580 .061 -.089 -.100 -.291 -.098 .269 -.016 .525
Over2 .665 .213 -.051 .170 .106 .004 .087 .039 .571 .213 -.177 -.089 -.189 -.157 .214 .154 .540
Cont3 .595 .123 .235 .184 -.006 -.006 -.065 -.033 .533 .044 -.123 -.193 -.272 .000 .206 -.092 .464
KSk9 .451 .168 .216 .173 .136 -.104 .248 -.028 .581 -.113 -.089 -.043 -.043 -.156 .115 .031 .400
I5S2 .078 .722 .023 .098 .129 -.017 .019 .050 .412 .294 -.306 -.039 .437 .031 -.066 -.102 .558
I5S4 -.012 .720 .074 .095 -.037 .065 .052 .060 .358 .219 -.304 -.074 .463 .211 -.098 -.055 .545
I5S3 .138 .671 .063 .026 .107 -.064 .026 .019 .401 .226 -.313 -.064 .403 .002 .030 -.115 .491
I5S1 .171 .669 .068 -.053 -.004 .067 -.021 -.007 .358 .264 -.277 -.145 .390 .137 .125 -.081 .489
I5S5 .189 .639 -.110 .155 -.031 .002 .076 -.013 .390 .279 -.352 -.160 .306 .034 -.062 .096 .487
KSk2 -.074 .005 .708 .111 .096 .068 .144 .043 .372 -.408 .230 .110 .084 .239 -.042 -.346 .556
KSk1 .201 .079 .702 .125 .043 -.024 .139 .048 .535 -.382 .062 .034 -.017 .184 .092 -.316 .580
KSk4 .165 .002 .700 .079 -.022 .049 .117 -.120 .437 -.420 .169 -.122 -.020 .205 .113 -.296 .554
KSk3 .100 .044 .666 -.004 -.039 -.022 .198 .010 .389 -.457 .075 .020 .065 .212 .138 -.250 .497
Sp2 .274 .027 .018 .775 .006 .050 .150 .041 .563 .030 -.050 -.076 -.345 -.009 -.478 .174 .704
Sp3 .172 .133 .086 .747 .107 .031 -.053 .016 .518 .136 -.047 -.091 -.261 -.007 -.508 -.066 .628
Sp1 .190 .158 .126 .675 .122 .012 .007 .050 .546 .094 -.060 -.041 -.208 -.010 -.437 -.061 .551
KSk8 .185 -.001 .135 .446 -.007 .051 .219 -.031 .501 -.234 .075 -.068 -.156 .085 -.221 .008 .396
VSI2 -.014 .033 .019 .129 .724 .094 .017 .046 .275 .252 .367 .292 .116 -.361 -.123 -.187 .553
VSI1 .056 .097 -.048 .107 .720 .048 .102 -.097 .313 .238 .334 .167 .187 -.469 -.086 -.102 .567
VSI4 .122 -.049 .148 .018 .685 .189 .009 .095 .328 .207 .440 .310 .028 -.256 .071 -.205 .553
VSI3 .110 .062 -.024 -.040 .642 .070 .069 .029 .258 .236 .290 .248 .144 -.370 .077 -.099 .441
SDO3 .019 .019 .061 .019 .188 .743 -.121 -.041 .150 .388 .565 -.056 .037 .331 .042 .023 .609
SDO1 -.082 .100 .049 .174 -.037 .721 .060 -.066 .180 .242 .444 -.133 .106 .451 -.145 .191 .578
SDO4 .105 .060 -.054 -.055 .131 .714 .053 .037 .179 .370 .464 .011 .080 .304 .133 .227 .552
SDO2 .075 -.111 .018 -.014 .107 .643 -.064 -.028 .087 .281 .508 -.043 -.069 .276 .096 .107 .448
KSk6 .011 .071 .185 .047 .094 .021 .787 .040 .406 -.438 .126 .212 .315 -.058 -.027 .389 .672
KSk5 .114 .029 .237 .044 .089 -.007 .776 -.025 .455 -.482 .135 .142 .247 -.097 .037 .365 .683
KSk7 .056 .032 .194 .065 .027 -.086 .775 -.002 .391 -.512 .048 .152 .248 -.101 -.020 .378 .655
WSR3 -.059 .080 -.044 -.010 .009 -.012 .007 .822 .033 .157 -.279 .712 -.067 .268 -.006 .032 .688
WSR1 .097 .094 .014 -.037 .054 -.016 -.024 .812 .138 .181 -.281 .686 -.119 .235 .109 -.015 .683
WSR2 .053 -.065 .007 .120 .004 -.060 .027 .770 .112 .062 -.243 .660 -.250 .202 -.042 .040 .618

Table 6.23 The factor loadings of measures for improvement outcomes from SPSS

The items to measure the use of shop floor management tools were loaded on to the four factors, which were developed originally, namely, Implementation of 5S practices (I5S), Use of the standard operations (SDO), Implementation of waste removal (WSR), and Use of visual management (VSI).

Nevertheless, the items to measure the improvement implementation were loaded differently. Improvement of Knowledge and Skills (KSk) were loaded separately onto two different factors. As articulated by Doolen et al. (2003), the items from KSK1 to KSK4 were originally developed to measure knowledge of improvement. These 4 items were renamed improvement knowledge (IpKn). The items from KSK5 to KSK7 were
originally developed to evaluate shop floor skills. Hence, were named shop floor skills
(SFK).
The 3 Improvement Contribution (Cont) items and the 2 Overall Improvement Perceptions (Over) items were loaded together. A further item from the measure of Improvement of Knowledge and Skills (KSk9) was also loaded onto this component. These 5 items were grouped together and given a new name: shop floor performance (SFP).

All 3 Sense of participation (Sp) items were loaded together into a single component. A further item from the measure of Improvement of Knowledge and Skills (KSk8) was loaded onto this component. These 4 items were clubbed together and given a new name: sense of participation (Sens).

The items to measure Improvement of Knowledge and Skills (KSk) were loaded separately onto two different components. As proposed by Doolen et al. (2003), the items from KSK1 to KSK4 were originally developed to measure knowledge of improvement. These 4 items were renamed improvement knowledge (IpKn). The items from KSK5 to KSK7 were originally developed to evaluate shop floor skills. They were named shop floor skills (SFK).
All 33 items were retained with high convergent validity (the items within the same scale are correlated, cross-loadings > 0.4) and discriminant validity (the items between different scales are distinct, cross-loading < 0.3) as defined by Hair et al.,(2010), Gaskin (2011) and Stangor (2011). In addition, the revised scales were rational and in line with previouse research (Doolen et al., 2003). The revised scales were listed as above (Table 6.24): shop floor performance (SFP); Implementation of 5S practices (I5S); improvement knowledge (IpKn); sense of participation (Sens); Use of visual management (VSI); Use of the standard operations (SDO); shop floor skills (SFK) and Implementation of waste removal (WSR).

Rotated loadings Eigenvalues % Of Variance Cumulative %
Shop floor performance (SFP) 5.469 16.572 16.572
Cont2 .714
Over1 .676
Cont1 .674
Over2 .665
Cont3 .595
KSk9 .451
Implementation of 5S practice (I5S) 2.632 7.977 24.549
I5S2 .722
I5S4 .720
I5S3 .671
I5S1 .669
I5S5 .639
Improvement knowledge (IpKn) 2.430 7.364 31.912
KSk2 .708
KSk1 .702
KSk4 .700
KSk3 .666
Sense of participation (Sens) 1.989 6.027 37.939
Sp2 .775
Sp3 .747
Sp1 .675
KSk8 .446
Use of visual management (VSI) 1.837 5.567 43.506
VSI2 .724
VSI1 .720
VSI3 .685
VSI4 .642
Use of the standard operations (SDO) 1.545 4.683 48.189
SDO3 .743
SDO1 .721
SDO4 .714
SDO2 .643
Shop floor skills (SFK) 1.277 3.870 52.059
KSk6 .787
KSk5 .776
KSk7 .775
Implementation of waste removal (WSR) 1.156 3.503 55.562
WSR3 .822
WSR1 .812
WSR2 .770
Table 6.24 The revised scales with factor loadings

Reliability of the Revised Scales
Having completed the factor analysis, it is pertinent to check the scales’ internal consistency to make certain that all the designed and developed questions ‘hang together’ and measure the underlying construct (Field, 2005). Cronbach’s (1951) alpha was calculated to ascertain the reliability of the factors. The following Table 6.25 presents the resulting values and the associated minimum inter-item correlation values generated by SPSS.

Nunnally and Bernstein (1994) suggested that Cronbach’s alpha coefficient as a scale should reach 0.7 or above to corroborate the internal consistency of the containing items. Nevertheless, if numbers of items in the scale are less than 10, then optimal mean inter-item correlation values that range from 0.2-0.4 are acceptable (Pallant, 2007). In this way, the scales to measure the use of shop floor management tools remained the same.

Factors Cronbach’s Alpha
SFP 0.768
I5S 0.745
IpKn 0.716
Sens 0.713
VSI 0.679 (* 0.287)
SDO 0.686 (* 0.291)
SFK 0.764
WSR 0.735
Table 6.25 Cronbach’s Alpha Values for Revised Survey Scales (* minimum inter-item correlation value)

The revised scales to measure improvement implementation are depicted in Table 6.26. Based on the factor analysis results, these possess high construct validity which implies that the questions actually measure what they are designed to measure,( Hair et al., 2010; Stangor, 2011).

Revised Scales Item List
Shop floor performance (SFP) • Over1: Overall, the performance of my improvement activities was a success in my company
• Over2: Overall, my improvement activities were vital in my company
• Cont1: My improvement activities have a positive effect on the shop floor area
• Cont2: This shop floor area improved measurably as a result of my improvement activities
• Cont3: My improvement activities have improved the performance of this shop floor area
• KSk9: Overall, the improvement activities helped me and my colleagues work together to improve performance
Shop floor skills
(SFK) • KSk5: I can communicate new ideas as a result of participation in improvement activities
• KSk6: I gained new production skills as a result of participation in improvement activities
• KSk7: In general, the participation in improvement activities motivated me to perform better
Sense of participation
(Sens) • Sp1: I like taking part in the current improvement activities
• Sp2: I would like to take part in the improvement activities in the future
• Sp3: In general, I am comfortable working with others to identify improvements on my shop floor area
• KSk8: Overall, the improvement activities increased my work interests
Improvement Knowledge (IpKn) • KSk1: Overall, the improvement activities increased my knowledge of what CI is
• KSk2: In general, the improvement activities increased my knowledge of how CI should be applied
• KSk3: Overall, the improvement activities increased my knowledge of the need for CI
• KSk4: In general, the improvement activities increased my knowledge of my role in CI
Table 6.26 The revised measures for improvement outcomes

Summary

This chapter elaborated the data collection and screening procedures. The data were collected from 10 Indo-Japanese automotive joint ventures. A questionnaire was derived from pretested and validated questions. It was distributed using the self-administered method. 1000 questionnaires were distributed of which 527 were returned. However, 25 (6.8%) contained missing values, so 502 were valid samples, giving a response rate of 50.0.2%. SPSS was used to assess the construct validity and summarise the patterns of the collected samples.

In the subsequent chapter, the theoretical model will be developed. Thereafter, structural equation modelling with path analysis will be used to analyse the data, shape the proposed theoretical model, and test the hypotheses.