Flynn el al. (1990), after modifying the work of Alreck and Settle (1985), provided a concise description of some potentially useful scales pertaining to the research in production management. The Likert-scale (1932) is the most frequently used among these scales (Aday and Cornelius, 2006; Saunders et al., 2007), since it is highly reliable measuring tool (Corbetta, 2003; McNabb, 2008). It was proposed by Likert in 1932 and met with considerable success, which is still continues (Corbetta, 2003). The Likert-scale can be used to quite effectively to measure the highly complex nature of the subject, because, it uses simple evaluations/judgements to capture responses and the intensity of opinion (Albaum, 1997).
The present study opted for the Likert-scale to measure respondents’ attitudes relating to continuous improvement. The current study used “bipolar scales” as suggested by Krosnick and Fabrigar (1997, p143) to measure respondents’ attitudes on improvement outcomes .These Bipolar scales reflect two opposing alternatives with a clear conceptual midpoint. On the other hand, “unipolar scales” (scales reflecting varying levels of some construct with no conceptual midpoint and with a zero point at one end) were used to measure respondents’ degree and utilisation of shop floor management tools. The original Likert-scale offered five-point categories/alternatives (“strongly approve”, “approve”, “undecided”, “disapprove” and “strongly disapprove”) (Likert, 1932), but later on the numbers of categories/alternatives were changed to seven-point (Corbetta, 2003). This study used a seven-point rather than a five-point Likert-scale to produce interval or interval-like data for later parametric statistics (i.e. factor analysis and structure equation modelling).
Operationalised measures for improvement implementation
This section elaborates the indicators used to measure implementation of continuous improvement. A few past studies (Schuring and Luijten, 2001; Muthiah and Huang, 2007; Gupta and Boyd, 2008) have introduced the use of one-off performance indicators (e.g., throughput, inventory and operations expenses) to measure the results of improvement (Imai, 1986; Imai, 1997), but they were unable to capture the most vital features of continuous improvement, the process (Marin-Garcia et al., 2008).
This present research measured the success of the continuous improvement on the basis of employees’ ways of doing things, like the number of ideas they submitted (e.g., Karlsson and Ahlstrom, 1996), implemented (e.g., Winfield, 1994; Baides and Moyano–Fuentes, 2012), and time they spent in developing ideas (e.g., Prado, 2001; Marin-Garcia et al., 2008).These indicators have been used universally to measure the process of continuous improvement (e.g., Santos and Powell, 2001; Seyedhosseini et al., 2011; Miina, 2012). Lillrank and Kano (1989) observed that the number of QCC meetings is a proxy to measure the quantity of QCCs, whilst the number of completed and presented QCCs is a proxy for quality of QCCs. The following questions were adopted to measure the implementation of QCCs (Table 6.1).
Constructs Questions and measuring method
(QCCs) This was measured by the total time spent on QCC meetings in a month, calculated by the product of the length of each meeting and the frequency of meeting on a monthly basis.
• QC_Meet_Times In general, how many times do you meet every month for QCC?
• QC_Met_Length In general, how long does each QCC meeting last?
(QCCs) This was measured by the numbers of completed QCC themes and the presentation made on the themes at company level.
• QC_Comp How many QCC themes did your group complete last year?
• QC_Pres How many times did you present in the meetings?
Table 6.1 The questions developed to measure the QCCs
The implementation of Teians was measured by the same method (Table 6.2). In consonance with a number of previous studies (Cusumano and Takeishi, 1991; Coleman, 1993a; Coleman, 1993b; Nihon HR Ky?kai, 1995; Jha et al., 1996; Terziovski and Sohal, 2000; e.g., Rapp and Eklund, 2002), the number of Teians submitted is a proxy to measure the quantity of Teians while the number of acceptances is a proxy to measure the quality of Teians.
Constructs Questions and measuring method
Quantity (Teians) This was measured by the number of submitted Teians.
• Tn_Sub How many Teians did you submit in the last 12 months?
Quality (Teians) This was measured by the number of accepted Teians.
• Tn_Acc How many of these Teians were accepted for implementation?
Table 6.2 The questions developed to measure the Teians
Operationalised measures for the use of the shop floor management tools
This part of questionnaire constitutes 4 scales (16 questions) to measure the degree of adoption and utilisation of the four building block shop floor management tools, complying CINET (2002), Soriano-Meier (2002), Rahman (2001) Terziovski and Sohal (2000). The four shop floor management tools were selected based on Bateman (2001), and Bateman and Rich (2003). The respondents were asked to evaluate the frequency of their use of these tools using seven-point Likert-scales, ranging from 1 (Never) to 7 (Always), and the sum of the answers was used as an indicator to uncover the utilisation of each building block tools (a higher score indicates greater utilisation of the tool) (Table 6.3).