Impact Of Iso-9000 Certification On Quality

TOTAL QUALITY MANAGEMENT, VOL. 13, NO. 1, 2002, 53- 67

Impact of ISO 9000 certi® cation on quality management practices: A comparative study Hesan A. Quazi,1 Chang Wing Hong2 & Chan Tuck Meng3 1

Division of Strategy, Management and Organizations, Nanyang Business School, Nanyang Technological University, Singapore 639798, 2Nanyang MBA Program, Nanyang Business School, Nanyang Technological University, Singapore, 639798 & 3Blk 288A, Bukit Batok St 25, #14- 224, Singapore 650288

abstract The primary objective of this study is to examine the relationship between ISO 9000 certi® cation and quality management practices and quality results of ® rms in Singapore. The ® ndings are compared with those reported by Rao et al. (1997, Total Quality Management, 8, pp. 335346) on ® rms in the USA, Mexico, China and India. Eight constructs for quality management practices are identi® ed. Using the questionnaire developed by Rao et al. a survey is carried out on selected ® rms from the manufacturing, service and construction industries in Singapore. Both ISO 9000 registered and non-registered ® rms are included in the survey. The questionnaire consists of 91 statements, on a ® ve-point Likert scale, to represent all the eight constructs. A total of 93 valid responses were received. The respondents are mainly from service industries, followed by construction and manufacturing. The majority of these ® rms are small and medium-sized enterprises, which are locally owned and have local customers and competitors. The data set is subjected to reliability and validity tests. Analysis of variance is carried out to explore the relationship between ISO 9000 certi® cation and quality management practices. The result reveals that the ISO certi® cation does not aþ ect quality management practices and quality results of ® rms in Singapore. Further analyses indicate that some organizational characteristics, such as size and industry, have relationships with certain constructs of quality management practices.

Introduction Objectives ISO 9000 is a family of standards that provides a series of guidelines on how to establish a quality system to manage the processes that aþ ect its product or services. This family of standards was ® rst published by the International Organization for Standardization (ISO) in 1987 and was subsequently updated in 1994 and 2000. The quality system is required to be documented and employees are expected to follow consistently the documented procedures. After the quality system is implemented, the ® rms can obtain registration through an audit performed by an independent (third-party) registrar. Correspondence: H. A. Quazi, Division of Strategy, Management and Organizations, Nanyang Business School, Nanyang Technological University, Singapore 639798. Tel: (65) 790 6347; Fax: (65) 7901 3697. E-mail: [email protected] ISSN 0954-4127 print/ISSN 1360-0613 online/02/010053-15 DOI: 10.1080/09544120120098564

© 2002 Taylor & Francis Ltd

54

H. A. QUAZI ET AL.

Rao et al. (1997) carried out a study to evaluate the eþ ects of ISO 9000 on quality management practices and quality results. Data were collected from ® rms in the USA, Mexico, India and China, which were then consolidated to form a global perspective. The present study replicates the original research done by these authors. The primary objective is to examine the relationship between ISO 9000 registration and quality management practices and quality results of ® rms in Singapore and to compare the ® ndings with those reported by Rao et al. (1997).

Importance/usefulness and potential contribution of research Proponents of ISO 9000 generally view the bene® ts of such registration as having the ability to improve product or service quality, eý ciency and productivity, customer con® dence, competitive advantage and the like. Criticism of ISO 9000 generally relates to the high level of paperwork and documentation, a rigid system that does not support creativity or empowerment, a standard that aims for consistency but not continuous quality improvement, etc. (Barnes, 1998; Larson, 1999). In the Singapore context, the trend of ISO 9000 registration is growing, and since 1989 more than 2500 ® rms have been registered to ISO 9000 standards (PSB, 1999). In addition, the government in Singapore is encouraging local industry to upgrade its operations and provides incentive to help defray the cost of the ISO 9000 implementation. Some large public sector ® rms are also subtly pressurizing their suppliers to register for ISO 9000 (BCA, 1999). The results of the proposed study will help to assess the relationship between ISO 9000 registration and quality management practices and quality results. This will also help business leaders, quality practitioners and governmental authority to determine the future path of ISO 9000 in Singapore and its relevance in enhancing the operations of a ® rm.

Literature review Eþ ects of ISO 9000 certi® cation In a study conducted in two manufacturing plants, one ISO 9000 registered and the other non-ISO 9000 registered, of a large corporation in the USA, it was found that the ISO 9000 quality programme improved the participants’ quality of work life (Elmuti & Kathawala, 1997). In addition, there was a positive impact on employee productivity and export sales. Chittenden et al. (1998 - 99) found that the UK ® rms that adopted ISO 9000 tended to be large, multi-product and manufacturing based. On the other hand, the ® rms that did not adopt ISO 9000 tended to be smaller businesses that dealt with domestic customers and served the local market. The study concluded that a high majority of ISO 9000 users felt that the advantages of using ISO 9000 outweighed the disadvantages. McAdam and McKeown (1999) showed that ISO 9000 certi® cation resulted in bene® ts. Better control of business, increased sales/business, reduced costs, increased productivity and fewer customer complaints were the bene® ts derived from ISO 9000 registration. The businesses that gained most from total quality management (TQM) had started with ISO 9000 and focused on external (e.g. customer satisfaction, etc.) as well as internal measures (scrap, eý ciency, etc.). Sun (1999) found that in Norwegian ® rms certi® cation of the ISO 9000 standards was signi® cantly correlated with quality results, especially the reduction of defective products and customer complaints, and business performance such as pro® tability and productivity.

ISO 9000 CERTIFICATION

55

However, the ISO 9000 certi® cation had little in¯ uence on market position and competitiveness, and no in¯ uence on employee satisfaction and environment protection. Anderson et al. (1999) found that manufacturing ® rms in North America adopted ISO 9000 as a means to provide credible signals of quality assurance to external parties. ISO 9000 complements rather than substitutes TQM. In addition, ISO 9000 was adopted as a tool in a grand strategy of achieving competitive advantage through quality management and communicating quality results. Simmons and White (1999) found that in North American registered ® rms did not realize advantages in operational performance and foreign sales over non-registered ® rms. However, in support of ISO 9000, it was found that the average ISO 9000-registered ® rms in the electronics industry were larger and more pro® table than the average non-ISO 9000 ® rms, although ® rm size did not aþ ect pro® tability. In Singapore, Quazi and Padibjo (1998) found that ISO 9000 registration provided a stepping stone toward TQM practices. The authors also reported that despite the bene® ts, there were numerous barriers faced by ® rms. Relationship between ISO 9000 and quality management practices The family of ISO 9000 standards has faced criticism that the certi® cation process fails to deal with some important aspects of TQM practices such as leadership, strategic planning and employee empowerment (Rao et al., 1997). Therefore, questions have been raised whether an ISO 9000 registration can result in an eþ ective implementation of quality management practices. It is argued that if the clauses of ISO 9001 are interpreted in depth and detail, they contain all the requirements that are associated with quality management practices. For example, clauses 4.1, 4.2 and 4.18 deal with management responsibility, quality system and training, respectively, which require practices that are in line with leadership, strategic planning and employee empowerment practices of quality management (Rao et al., 1997). The literature indicates that there is a link between the bene® ts reported by ISO 9000registered ® rms and the constructs of quality management practices listed by Rao et al. (1997). These constructs are leadership, information and analysis, strategic quality planning, human resource development, quality assurance, supplier relationships, customer orientation and quality results. A summary of the pertinent literature on the eight constructs of quality management practices is shown in Table 1. Theoretical framework and research hypothesis As this is a replicative study the theoretical framework and the methodology are similar to those of Rao et al. (1997), but the measurements have been slightly amended to suit the local context and to align it with the Singapore Quality Award criteria. For the data analyses the approach adopted by Saraph et al. (1989) and Steeples (1992 - 93) were followed. Conceptualization of the constructs Based on the literature in the ® eld, Rao et al. (1997) conceptualized the quality management constructs listed above, which are presented below: ·

Leadership: The leadership construct is conceptualized by senior management’s personal involvement, acceptance of responsibility, visibility and shared vision and goals.

56

H. A. QUAZI ET AL.

Table 1. Summary of relationship between literature review ® ndings and constructs of quality management practices Constructs of quality management practices

Literature review ® ndings

1 Leadership · · ·

2 Information and analysis ·

·

·

3 Strategic quality planning 4 Human resource development ·

·

·

·

5 Quality assurance ·

·

6 Supplier relationship ·

7 Customer orientation · · · ·

8 Quality results

·

·

·

·

·

·

·

·

·

·

·

·

·

·

·

·

·

Many ® rms are not addressing or implementing TQM methodology properly (Barclay, 1993) Employee involvement in planning, decision-making and corporate strategy was limited (Barclay, 1993) Failure of QSA and TQM programmes seemed to be due to implementation problems and non-committal of executives rather than from a faulty model of TQM (Barclay, 1993) Without accurate and timely information, quality control is impossible (Garvin, 1988) Remedial actions require an understanding of problems and their causes (Garvin, 1988) Data must be in the right hands otherwise the problems may not be solved because authority is lacking or communication is inadequate (Garvin, 1988) Strong belief that continuous improvements is a strategic objective (Barclay, 1993) ISO 9000 quality programme improves participantsp quality of work life (Elmuti & Kathawala, 1977) Reduce cost of training (Chittenden et al., 1998- 99) Reduce supervision (Chittenden et al., 1998- 99) Streamlined procedures and documentation (Quazi & Padibjo, 1998) Increased consciousness for preventive and corrective actions (Quazi & Padibjo, 1998) Organizations without formal programmes to prevent degradation of supplier quality had higher supply rejection rates (Simmons & White, 1999) · Fewer complaints (Chittenden et al., 1998- 99) Retained business (Chittenden et al., 1998- 99) Increased customer preference (Quazi & Padibjo, 1998) Improved quality image (Quazi & Padibjo, 1998) Compliance to customer requirements (Quazi & Padibjo, 1998) Positive impact of employee productivity (Elmuti & Kathawala, 1977) Increase percentage of time spent on actual production (Elmuti & Kathawala, 1977) Produced a higher rate of products (Elmuti & Kathawala, 1977) Increased export sales (Elmuti & Kathawala, 1977) Quality of products (Elmuti & Kathawala, 1977) Cost savings and indirect bene® ts were greater than the cost of running the programme (Elmuti & Kathawala, 1977) ISO quality programme believed to help enhance morale, decrease employee absenteeism, contribute to organizational goals of increased productivity, export sales, reduced costs and improved quality of work (Elmuti & Kathawala, 1977) Believed that these bene® ts would continue to accrue for the foreseeable future (Elmuti & Kathawala, 1977) Manufacturing ® rms that implement ISO 9000 are primarily motivated by the desire to improve internal processes while small ® rms are motivated by marketing and competitive advantages (Chittenden et al., 1998- 99) Reduce scrap waste (Chittenden et al., 1998- 99) Capacity utilization (Chittenden et al., 1998- 99) New markets (Chittenden et al., 1998- 99) Increased sales (Chittenden et al., 1998- 99) Reduced scrap/down time (Chittenden et al., 1998- 99) Increased/wider sales (Chittenden et al., 1998- 99) Increased pro® t (Chittenden et al., 1998- 99) Improved competitiveness in market (Quazi & Padibjo, 1998)

ISO 9000 CERTIFICATION

·

57

·

Information and analysis: The information and analysis construct is conceptualized by the availabilit y of data, timeliness of data and usage of data. Strategic planning: The strategic quality planning construct is conceptualized by the integration of quality management and customer satisfaction in organizational plans, long-term quality vision and deployment and understanding of quality goals. Human resource development: The human resource development construct is conceptualized by continuous training and education, empowerment and providing resources and conducive environment. Quality assurance: The quality assurance construct is conceptualized by new product design review procedures, design for manufacturability procedures, control of speci® cation and procedures, preventive maintenance and quality control activities along the value added chain. Supplier relationship: The supplier relationship construct is conceptualized by supplier selection criteria, number of suppliers, exchange of information and services, supplier involvement in new product development and duration of relationship. Customer orientation: The customer orientation construct is conceptualized by commitment to satisfy customers, integration of customer satisfaction in ® rm’s goals and vision, knowledge of customer needs and expectations, usage of customer feedback in new product design, monitoring customer satisfaction, responsiveness to customer complaints and level of interaction with customers. Quality results: The quality results construct is conceptualized by levels of scrap and rework, throughput time, warranty costs, customer complaints, productivity, pro® tability, market share, costs and competitive position. ·

·

·

·

·

Hypotheses Based on the literature review, the following hypotheses have been developed for this study: H1: ISO 9000 quality management system has a positive impact on leadership of an organization. H2: ISO 9000 quality management system has a positive impact on information and analysis of an organization. H3: ISO 9000 quality management system has a positive impact on strategic quality planning of an organization. H4: ISO 9000 quality management system has a positive impact on human resource planning of an organization. H5: ISO 9000 quality management system has a positive impact on quality assurance of an organization. H6: ISO 9000 quality management system has a positive impact on supplier relationship of an organization. H7: ISO 9000 quality management system has a positive impact on customer orientation of an organization. H8: ISO 9000 quality management system has a positive impact on quality results of an organization. The independent variable in this study is the ISO 9000 registration status of the ® rms (i.e. ISO 9000-registered and not registered). The dependent variables are the constructs of quality management practices. Eight factors pertaining to quality management practices and quality results that were identi® ed earlier in this paper formed the basis for the constructs of the dependent variables (listed in the section `Conceptualization of constructs’ ).

58

H. A. QUAZI ET AL.

Research methodology Questionnaire design As this study is intended to replicate the original research done by Rao et al. (1997), the questionnaire used was largely the same. However, some minor modi® cations and additions were made that were necessary in order to structure the questionnaire to suit the Singapore context and to align it to the Singapore Quality Award criteria. The questionnaire consists of eight factors of quality management practices and ® ve to 16 statements for each factor. A ® ve-point Likert scale, with a provision for `Unable to respond’ , was used (where 5 5 very high and 1 5 very low). In addition, there are also questions to obtain the pro® les of the ® rms and respondents. These were generally closeended, multiple-choice type questions. Nature of sample Samples were chosen from manufacturing, service and construction industries. In each industry, the samples were divided into two groups. Group 1 consisted of ® rms that are already ISO 9000 registered, while group 2 consisted of ® rms that are not ISO 9000 registered. Samples were chosen randomly from a number of directories (e.g. the Singapore business directoriesÐ Singapore Electrical Products & Services, Singapore Construction Services & Products, Singapore Business Services, Singapore Electronics Trade, Singapore Trade Logistics Services, Singapore Supporting Industries, PSB Directory of Certi® ed Products and Companies, BCA Directory of Registered Contractors, Singapore Top 1000 Companies and Green Book Directory). Each page of the directories was assigned with a unique running number and then a random number generator was used to generate numbers to select the pages. All companies listed in the selected pages were taken as the sample. In general, the sample size should be adequate to enable generalization or projection of the sample characteristics to that of the population characteristics. For this study, a target sample size of 100 was set. It was expected that a minimum sample size of 30 for each group should be obtained, as the minimum acceptable sample size for correlational research or causal-comparative research is 30 (Gray & Deihl, 1992). Further, Chao (1980, pp. 227, 322) suggests that a minimum sample size of 30 is suý cient to ensure normal distribution of sample. A total of 460 questionnaires were sent out and 93 valid returns were received, representing a response rate of about 20%. Outliers and invalid returns were excluded by a visual scrutiny and checking process by the researchers. Most of the returns were received between 1 and 3 weeks after sending. Method of data collection A mail/fax survey was used for this study. Various trade directories and other related sources were used to compile a list of the 3100 target ® rms. A one-page letter was ® rst faxed to 3100 ® rms to determine the status of their ISO 9000 registration prior to sending the actual survey. Depending on the response to the pre-survey letter, the researchers were able to classify the ® rms belonging to either group 1 (i.e. ISO 9000 registered) or group 2 (i.e. not ISO 9000 registered). Questionnaires were then sent (by mail, fax and by hand) to 460 ® rms. Each questionnaire was accompanied by a cover letter explaining the purpose of the survey. Questionnaires that were sent by mail and hand had a self-addressed postage pre-paid envelope enclosed.

ISO 9000 CERTIFICATION

59

Each ® rm was requested to respond within 2 weeks of receiving the questionnaire. The response to the survey was closely monitored, and telephone calls were made to ® rms that did not respond by the stipulated time frame. The one-page pre-survey letter was addressed to the Chief Executive Oý cer (CEO) of the ® rm. The questionnaires were addressed to the person who responded to the pre-survey letter. The targeted respondents were the Chief Executives or their direct reports, who have responsibility for quality management. This was done to ensure that the respondents had the organizational knowledge to answer the questions correctly. Most of the responses that were received were either from the CEO, Director, Quality Manager or the management representative of the ISO 9000 system. Data analysis To establish the credibility of the instrument, reliability and validity tests were performed. The reliability was measured by Cronbach’s alpha. This ensured repeatability of the results when diþ erent people took the test or tested on diþ erent occasions. Content validity was judged subjectively by the researchers, criterion-related validity was measured by multiple regression coeý cient and construct validity was examined by factor analysis. Analysis of variance (ANOVA) was used to explore the relationship between ISO 9000 certi® cation and quality management practices. The F-value was used to determine whether to accept or reject the null hypothesis (`The mean scores of group 1 and group 2 for each construct are equal). Eight one-way ANOVA tests were conducted, one for each construct. Analysis and ® ndings Pro® le of respondents There was a total of 93 respondents consisting of 59 ISO 9000-registered ® rms and 34 nonISO 9000-registered ® rms. Of these, 14 were members of the Singapore Quality Class (The Singapore Quality Class Companies are the ® rms which have been assessed by the Singapore Productivity and Standards Board to be potential applicants for the Singapore Quality Award in the future) and eight had submitted applications for the Singapore Quality Award. Although some of the questionnaires were not fully completed with regards to the ® rms’ particulars, it was checked that all essential information was present for the purpose and the responses were judged to be usable. Questions that were unanswered by respondents were treated as missing data and were excluded from the analysis. Industry classi® cation. About 28% of the sample were from manufacturing, 38% from service and the rest, 34%, from the construction industry. Of the total sample, 63% were ISO 9001/ 9002 certi® ed. The proportions of ISO-certi® ed companies in manufacturing, service and construction sectors were 88, 49 and 59%, respectively. Organization size. About 63% of the sample come from organizations with fewer than 150 employees. Of this group of companies, 51% were ISO 9001/9002 certi® ed. Further, of the total sample, about 43% had annual sales of up to S$10.00 million and about 9% had above S$500 million. Customer pro® le. More than half of the sample ® rms (54%) reported having local customers, whereas only 22% of them had a global customer base. About 80% of the ® rms that had a

60

H. A. QUAZI ET AL.

Table 2. Descriptive statistics Constructs

Survey questionnaire number

Mean

SD

Leadership Information and analysis Strategic quality planning Human resource development Quality assurance Supplier relationship Customer orientation

1, 5, 33, 34, 40, 52, 61, 64 7, 10, 12, 14, 16, 38, 44, 48 20, 21, 23, 25, 36, 45, 51 9, 13, 27, 30, 31, 35, 37, 60, 8, 11, 18, 32, 39, 43, 46, 47, 3, 17, 50, 53, 55, 57, 59, 69, 4, 24, 26, 29, 42, 58, 65, 71, 81, 82, 85, 86, 89 2, 19, 28, 54, 56, 62, 66, 67,

3.7842 3.4059 3.6224 3.2410 3.6510 3.3620

0.1285 0.1404 0.1664 0.2985 0.1673 0.2720

3.5849 3.4474

0.2755 0.1281

Quality results

68, 49, 77, 76,

75, 83 84, 88 87 78, 79, 80,

70, 73, 74

global customer base were ISO 9001/9002 certi® ed compared with only 60% of those that had a local customer base. Competitor pro® le. About half of the sample ® rms reported having local competitors, whereas 22% of them had global competition. Of those ® rms which had global competitors, 77% were ISO 9001/9002 certi® ed as compared with only 62% of those who had reported a local competitor base. Ownership. More than half (54%) of the sample ® rms were 100% locally owned and about one-third (30%) were 100% foreign owned. The rest of the ® rms had various other types of ownership. Of the locally owned ® rms, 55% were ISO 9001/9002 certi® ed as compared with 78% of the foreign owned companies.

Descriptive statistics Of the eight constructs of quality management practices, `leadership’ shows the highest mean score (3.7842) and `human resource development’ shows the lowest mean score (3.2410) on a ® ve-point Likert scale. The quality management constructs, number of items measuring each construct, and the means and standard deviation of each of the constructs are shown in Table 2.

Reliability of data The reliability of empirical measurements is indicated by the internal consistency, which can be estimated by using a reliability coeý cient such as Cronbach’ s alpha. An internal consistency analysis was performed separately for each of the constructs of quality management. Table 3 shows the Cronbach’ s alpha corresponding to each of the constructs. Typically, a reliability coeý cient of 0.7 (Nunally, 1967) or higher is considered to be adequate. Higher values of Cronbach’s alpha indicate a higher degree of reliability. The reliability values for this study range from 0.8708 to 0.9549. Accordingly, the results of this survey provide strong evidence that the instrument is reliable. Cronbach’s alpha values of Rao et al.’s study are also shown in Table 3. It can be seen from the table that the alpha values reported by these authors are also very high and greater than or equal to 0.83 for all the constructs.

ISO 9000 CERTIFICATION

61

Table 3. Test for internal consistency Cronbach’s alpha Constructs Leadership Information and analysis Strategic quality planning Human resource development Quality assurance Supplier relationship Customer orientation Quality results

This study

Rao et al.’s study

0.9051 0.9265 0.9328 0.9141 0.9282 0.8708 0.9549 0.9342

0.83 0.87 0.86 0.88 0.91 0.92 0.91 0.89

Validity of the instrument The validity of a measure refers to the extent to which it measures what it is intended to measure. Three diþ erent types of validity are generally considered: ·

content validity; criterion-related validity; · construct validity. ·

It is to be noted that Rao et al. (1997) did not report the construct and criterion-related validity data and as such no comparison with those of the present study is possible. Content validity. Content validity depends on how well the researchers created the measurement items to cover the content domain of the variables being measured (Nunally, 1967). A measure has content validity if there is a general agreement among the subjects and researchers that the instrument has measurement items that cover all aspects of the variable being measured. It is not evaluated numerically but the researchers subjectively judge it. In this study, the questions in the survey were mainly based on a previous study reported by Rao et al. (1997), who had selected the measurement items based on an extensive literature review. Furthermore, the authors of this study reviewed the relevant literature and examined the contents of the Singapore Quality Award (SQA) and made some minor adjustments to the instrument. Therefore, it is judged that there is content validity. Criterion-related validity. Criterion-related validity, sometimes called predictive validity or external validity, is concerned with the extent to which a measuring instrument is related to an independent measure of the relevant criterion. Therefore, the measures of quality management have criterion-related validity if these constructs collectively are highly and positively correlated with a measure of organization quality performance (Saraph et al., 1989). Organizations’ quality performance is a dependent variable and is measured by scrap levels, customer complaints, competitive position of organization, throughput time of product/ service, various costs in organization, customer satisfaction and similar outcome measures. This study has evaluated the criterion-related validity of the combined set of seven constructs of quality management practices by examining the multiple correlation coeý cient computed for these seven constructs (independent variables) and the quality result construct (dependent variable). The quality result construct measures the organizational outcomes in terms of reduction of warranty costs, rework levels, scrap levels and throughput time, increase

62

H. A. QUAZI ET AL.

Table 4. Summary of factor matrix for each construct

Construct

Item loading range for factor 1

Eigenvalue

% Variation explained by factor 1

Leadership Information and analysis Strategic quality planning Human resource development a Quality assurancea Supplier relationshipa Customer orientationa Quality results

0.734- 0.817 0.759- 0.860 0.749- 0.889 0.475- 0.790 0.594- 0.850 0.563- 0.851 0.365- 0.840 0.566- 0.825

4.839 5.349 5.009 6.074 6.509 4.894 10.011 6.68

60.492 66.866 71.552 55.222 59.171 48.942 58.886 60.725

a

Two factors extracted.

in productivity and the like. The multiple correlation coeý cient of the seven constructs and the quality result is found to be 0.959 (F 5 48.73, pÜ 0.000). Therefore, it is concluded that the quality management constructs have an acceptabe degree of criterion-related validity when taken together. Construct validity. A measure has construct validity if it measures the theoretical construct that it was designed to measure. The construct validity of each construct was evaluated by factor analysing the measurement items of each of the factors. A rotation procedure, Varimax, was applied in order to maximize the correlation of each item on a factor. The results of the factor analysis are shown in Table 4. The factor matrices showed that four of the factors (constructs) were uni-factorial and the other four were bi-factorial. Human resource development, quality assurance, supplier relationship and customer orientation were the constructs that emerged as bi-factorial. Therefore, a total of 12 factors emerged from factor analysis. It appears that there could be a total of 12 factors, rather than eight. Alternatively, the 12 factors could be a combination of eight factors and four sub-factors (see Quazi et al., 1998; Saraph et al., 1989). As the development of scales for quality management practices is not the primary focus of the study, the issue is not explored any further. Analysis of variance ANOVA was used to explore the relationship of ISO 9000 registration and quality management practices. Eight one-way ANOVA tests were conducted, one for each construct. The mean value for the given construct was taken as the dependent variable. The non-metric independent variable was the status of ISO 9000 registration of the respondents, either ISO 9000 registered or non-ISO 9000 registered. The F-ratios (ranging between 0.016 and 3.308) indicate that the diþ erences between these two groups of ® rms are not statistically signi® cant at the 5% level (p-values ranging between 0.075 and 0.898). The results therefore, do not provide support that ISO 9000 registration has a positive impact on quality management practices. Discussion From the Singapore sample there is no statistically signi® cant relationship between ISO 9000 registration status and quality management practices and quality results of ® rms. This

ISO 9000 CERTIFICATION

63

conclusion is diþ erent from that of Rao et al. (1997) who found that ISO certi® cation did aþ ect quality management practices and quality results. To identify possible reasons for such disagreement, ® ndings of other studies focusing on organizational characteristics and other possible intervening variables are examined in the following sections. A study conducted by Terziovski and Samson (1999) in Australia and New Zealand on manufacturing ® rms reported that there were signi® cant diþ erences in the relationship between TQM and organizational performance across industry sectors and size of the ® rms, industry type and ownership. Ismail and Hashmi (1999) and Kie and Palmer (1999) also found that the adoption of TQM and its corresponding eþ ectiveness were dependent on site size/number of employees. Wiele and Brown (1997/1998), in their study in Australia, found that most small and medium-sized enterprises (SMEs) seemingly felt forced by their customers to go for ISO 9000 certi® cation and did not move further down the quality path. Goh and Ridgway (1994) reported very similar ® ndings on SMEs in the UK. In the study of Rao et al. (1997), about 77% of the responses came from the manufacturing industries (500 out of 649). Moreover, their sample did not include the construction industry. The present study, however, has more respondents from the service and construction industries (i.e. 38 and 34%, respectively) than from the manufacturing industry. Therefore, it may be possible that in the case of Singapore the eþ ects of ISO may be industry sensitive. It is therefore very likely that since the industries selected in the present study are not identical to that of the study of Rao et al., a diþ erent result may be possible. It is likely that the conclusions drawn by Rao et al. may be more applicable speci® cally to the manufacturing industry, but not to all industries in general. In terms of size eþ ect, 30% of respondents of the Rao et al. study came from big ® rms with annual sales higher than US$500 million, compared with 10% for this study. On the other hand, the majority of respondents of the present study (62%) are from SMEs. This may mean that ISO 9000 registration is more eþ ective in improving quality management practices for large ® rms, and not so eþ ective for small ones. Based on the above discussion, it is possible that the diþ erence in results for this study compared with that of Rao et al. might be explained mainly by industry sensitiveness and ® rm size, i.e. ISO 9000 certi® cation may be more likely to have a positive eþ ect on the quality management practices of ® rms in certain industries. The eþ ect of ISO 9000 on quality management practices is also more likely to be positive for larger ® rms rather than the SMEs. In view of the above discussions it was decided to verify the assertions that: ® rm size has an impact on the level of quality management practices; and manufacturing ® rms are more likely to use ISO 9000 registration as the starting point towards the journey for TQM than the non-manufacturing ® rms. Further statistical tests were carried out and the results are discussed below. The eþ ect of ® rm size on quality management practices ANOVA tests were carried out to evaluate the impact of ® rm size on quality management practices. Two proxies for ® rm size were used, i.e. number of employees and sales revenue per year. The proxy served as the independent variables in the ANOVA test. In using number of employees as the proxy, the independent variables are `fewer than 150 employees’ and `equal or more than 150 employees’ . In using sales revenues as the proxy, the independent variables are `sales revenue less than US$25 million’ and `sales revenue more than US$25 million’ .

64

H. A. QUAZI ET AL.

ANOVA was used to test the following hypotheses: HE1: Firm size (represented by number of employees) regardless of its ISO 9000 registration status has an impact on quality management practices. HE2: Size (represented by number of employees) of ISO 9000-registered ® rms has an impact on quality management practices. HE3: Size (represented by number of employees) of non-ISO 9000-registered ® rms has an impact on quality management practices. HS1: Firm size (represented by sales revenue) regardless of its ISO 9000 registration status has an impact on quality management practices. HS2: Size (represented by sales revenue) of ISO 9000-registered ® rms has an impact on quality management practices. HS3: Size (represented by sales revenue) of non-ISO 9000-registered ® rms has an impact on quality management practices. The result of the ANOVA test between the organizational size, as measured by number of employees, and quality management practices indicates that, at 5% signi® cance level, it appears that ® rm size represented by the number of employees does not have an impact on quality management practices regardless of whether the ® rm is ISO registered or not (i.e. hypotheses HE1, HE2 and HE3 are not supported). However, among registered ® rms, the results do support HS2 (i.e. size of ISO 9000-registered ® rms has an impact on quality management practices) in several dimensions. ISO 9000-registered ® rms show signi® cant diþ erences in their quality management practices in the following dimensions: CON 4: Human resource development and management ( p 5 0.042). CON 5: Quality assurance of products and services ( p 5 0.011). · CON 8: Customer focus and satisfaction (p 5 0.029). · ·

The eþ ect of industry type on quality management practices Similarly, ANOVA tests were carried out to examine whether `industry type’ has any impact on quality management practices. In this case the independent variables are: (1) manufacturing; (2) service; and (3) construction industries. Hypotheses: HI1: Industry type, regardless of ISO registration status has an impact on quality management practices. HI2: Industry type of ISO-registered ® rms has an impact on quality management practices. HI3: Industry type of non-registered ® rms has an impact on quality management practices. As there are three diþ erent `industry types’ (i.e. manufacturing, service and construction), tests were carried out to determine which pair of industry groups has an impact on quality management practices. Three one-way ANOVAs were performed with the combined sample (i.e. including both ISO-registered and non-registered ® rms) and `only ISO-registered ® rms’ . In each of these samples, the paired groups were as follows: ·

Groups 1 and 2. Groups 1 and 3. · Groups 2 and 3. ·

ISO 9000 CERTIFICATION

65

Combined sample: ISO 9000-registered and non-registered ® rms The results show that manufacturing (group 1) and service (group 2) have diþ erences in quality management practices in the following dimensions: CON 2: Information and analysis (p 5 0.013). CON 5: Quality assurance of products and services ( p 5 0.009). · ·

Between groups 1 and 3 (i.e. manufacturing and construction), there appear to be signi® cant diþ erences in quality management practices in the following dimensions: CON 2: Information and analysis (p 5 0.007). CON 5: Quality assurance of products and services ( p 5 0.038). · CON 7: Quality results (p 5 0.017). · ·

Between groups 2 and 3 (i.e. service and construction), it appears that there are no diþ erences in quality management practices. Only ISO-registered ® rms Between groups 1 and 2 (i.e. manufacturing and service), it appears there is a signi® cant diþ erence ( p 5 0.042) in quality management practices along the dimension of `information and analysis’ only. Between groups 1 and 3 (i.e. manufacturing and construction), it appears there are signi® cant diþ erences in quality management practice along the dimensions of `information and analysis’ and `supplier quality’ (p 5 0.006 and 0.043, respectively). However, between groups 2 and 3 (i.e. service and construction) there are no signi® cant diþ erences in quality management practices. Conclusion This study suggests that in general ISO 9000 registration does not have any impact on quality management practices and quality results for ® rms in Singapore. The study further concludes that the diþ erences of quality management practices may be in¯ uenced by ® rm characteristics (i.e. size and industry). Singapore data covering the combined sample, represented by both ISO 9000-registered and non-registered ® rms, indicate signi® cant diþ erences of quality management practices between industry groups in terms of certain dimensions of quality management practices. Similar diþ erences are also found when only the ISO 9000-registered ® rms are analysed. However, no signi® cant impact of ® rm size on quality management practices (when measured by number of employees) is observed. Although the study of Rao et al. (1997) reports signi® cant diþ erences between ISO 9000 registered and non-registered ® rms in terms of quality management practices, it is suspected that the composition and characteristics of the ® rms in the sample may be the reason for such diþ erences in ® ndings. Further studies are necessary to establish these facts more strongly. Direction for future research and limitations Direction for future research It is recommended that future research be carried out on this subject by conducting a survey speci® cally on manufacturing industry in Singapore to determine if ISO 9000 registration has an impact on quality management practices and quality results as reported by Rao et al.

66

H. A. QUAZI ET AL.

(1997). If the results indicate that such registration does impact on quality management practices and quality results, then the study can be extended further to other industries to determine if the results are similar. Surveys may also be conducted speci® cally on SMEs and large organizations to determine the impact of size on quality management practices and quality results.

Limitations There are some limitations of this study that should be noted. Firstly, the sample size of this research (93 respondents) is comparatively smaller than that of Rao et al., which had 649 respondents. Furthermore, the response rate of 20% appears to be low, which may also aþ ect the outcome of the research. Secondly, 64% of the respondents in this study are from ISO 9000-registered ® rms and only 36% from non-ISO-registered ® rms. The relative imbalance between the number of respondents from the two groups may be a contributing factor resulting in diþ erent conclusions from those of Rao et al. Thirdly, the industries studied are con® ned to manufacturing, service and construction industries only. Fourthly, the data collected in this study are self-reported by individuals from the responding organizations and may have re¯ ected the personal perception and opinion of these individuals. Furthermore, the study is limited to a very small country where there is strong intervention by the government. Such intervention may have in¯ uenced the quality management practices in the sample organizations. Therefore, the ® ndings of this study should be interpreted carefully.

References Anderson, S.W., Daly, J.D. & Johnson, M.F. (1999) Why ® rms seek ISO 9000 certi® cation: Regulatory compliance or competitive advantage?, Production and Operations Management, 8, No. 1, pp. 28- 43. Barclay, C.A. (1993) Quality strategy and TQM policies: empirical evidence, Management International Review, 33, Special Issue 1, pp. 87- 98. Barnes, F.C. (1998) ISO 9000 Myth and Reality, a reasonable approach to ISO 9000, SAM Advanced Management Journal, 63, No. 2, pp. 23- 30. BCA (1999) ISO 9000 Certi® cation Scheme (Singapore, Building Control Authority). Chao, L.L. (1980) Statistics for Management (Monterey, CA, Brooks/Cole). Chittenden, F., Poutziouris, P. & Muhktar, S.M. (1998) Small ® rms and the ISO 9000 approach to quality management, International Small Business Journal, 17, No. 1, pp. 73- 88. Crosby, P.B. (1979) Quality is Free (New York, New American Library). Deming, W.E. (1986) Out of the Crisis (Cambridge, MA, MIT Centre for Advanced Engineering). Elmuti, D. & Kathawala, Y. (1997) An investigation into the eþ ects of ISO 9000 on participants’ attitudes and job performance, Production and Inventory Management Journal, 38, No. 1, pp. 52- 57. Everette, E.A., Hershauer, C.J. & Ruch, W.A. (1981) Productivity and Quality Measurement as a Basis for Improvement (Englewood Cliþ s, NJ, Prentice Hall). Frank, C. & Barnes, S.A.M. (1988) ISO 9000 myth and reality: a reasonable approach to ISO 9000, Advanced Management Journal, Spring. Garvin, D.A. (1988) Managing Quality (London, Collier Macmillan). Garvin, D.A. (1984) Japanese quality management, Columbia Journal of World Business, 19, pp. 3- 12. Garvin, D.A. (1983) Quality on the line, Harvard Business Review, 61, pp. 65- 75. Goh, P.L. & Ridgway, K. (1994) The implementation of TQM in small and medium-sized manufacturing ® rms, The TQM Magazine, 6, pp. 54- 60. Gray, L.R. & Deihl, P.L. (1992) Research Methods for Business and Management (New York, Macmillan). Ishikawa, K. (1976) Guide to Quality Control (Tokyo, Asian Productivity Organisation). Ismail, M.Y. & Hashmi, M.S.L. (1999) The state of quality management in the Irish manufacturing industry, Total Quality Management, 10, pp. 853- 862. Juran, J.M. (1981a) Product qualityÐ a prescription for the West, Part 1, Management Review, 70, pp. 8- 14. Juran, J.M. (1981b) Product qualityÐ a prescription for the West, Part 2, Management Review, 70, pp. 57- 61.

ISO 9000 CERTIFICATION

67

Kie, K.S. & Palmer, E. (1999) An empirical examination of ISO 9000-registered ® rms in New Zealand, Total Quality Management, 10, pp. 887- 899. Larson, M. (1999) Long-term improvement of fad? Where do we go from here?, Quality, January. Leonard, S.F. (1982) The incline of quality, Harvard Business Review, 60, pp. 163- 171. McAdam, R. & McKeown, M. (1999) Life after ISO: an analysis of the impact of ISO 9000 and total quality management on small businesses in Northern Ireland, Total Quality Management, 10, pp. 229- 241. Monden, Y. (1982) Toyota Production SystemÐ Practical Approach to Production Management (New York, American Institute of Industrial Engineers). Nunnally, J.C. (1967) Psychometric Theory (New York, McGraw-Hill). PSB (1999) PSB Awards over 1000 ISO 9000 Certi® cates (Singapore Productivity and Standards Board). Quazi, H.A. & Padibjo, S.R. (1998) A journey toward total quality management through ISO 9000 certi® cationÐ a study on small and medium sized enterprises in Singapore, International Journal of Quality & Reliability Management, 15, pp. 489- 508. Rao, S.S., Ragu-Nathan, T.S. & Solis, L.E. (1997) Does ISO have an eþ ect on quality management practices? An international empirical study, Total Quality Management, 8, pp. 335- 346. Saraph, J.V., Benson, P.G. & Schroeder, R.G. (1989) An instrument for measuring the critical factors of quality management, Decision Sciences, 20, pp. 810- 829. Simmons, B.L. & White, M.A. (1999) The relationship between ISO 9000 and business performance: Does registration really matter?, Journal of Managerial Issues, 11, pp. 330- 343. Steeples, M.M. (1992/1993) The Corporate Guide to the Malcolm Baldrige National Quality Award (Business One Irwin). Sun, H. (1999) Diþ usion and contribution of the total quality management: an empirical study in Norway, Total Quality Management, 10, pp. 901- 914. Terziovski, M. & Samson, D. (1999) The link between total quality management practice and organisational performance, International Journal of Quality & Reliability Management, 16, pp. 226- 237. Wiele, T.V.D. & Brown, A. (1997- 98) Venturing down the TQM path for SME’s, International Small Business Jour nal, 16, pp. 50- 68.