Lesser Used Aggregate Performance Indicators

Published Date: 02 Nov 2017

GDP can be calculated in three different ways: using the nations’ output, adding up the rewards to the factors of production, or calculating the total expenditure on both produced and consumer goods (Roussos, 1988). GDP can be represented in a number of ways, notably in terms of nominal value or in terms of constant prices. For the purposes of determining real growth or decline in performance, constant price GDP is most favoured. Constant price GDP allows comparison of GDP over an extended period of time, exchange rates and other factors against those of a base year or specific reference period.

Lesser Used Aggregate Performance Indicators

While capacity utilisation, GDP and employment statistics are arguably the most widely used aggregate performance indicators used to determine and report on the performance of the manufacturing sector, several other indicators are used for similar purposes.

The general acceptance of metrics apart from capacity utilisation, GDP and employment statistics is subject to various considerations among which are the source of the information and the ability of the stakeholders to derive meaning from the indicators.

Some important indicators which relate to the aggregate performance of the manufacturing sector include power consumption, volume of goods produced, factory space utilisation and manufacturing value added. Many more indicators are used in various subsectors, with some being product specific, and others being specialised to the type of processes undertaken in the manufacturing process. The lack of widespread utilisation of these metrics disqualifies them from analysis owing to the fact that their accuracy and adequacy is less likely to have grave consequences on the overall performance of the entire manufacturing sector.

Users And Uses Of Aggregate Performance Indicators

Some kind of performance measurement is a prerequisite for judging whether an operation is good, bad or indifferent. Without performance measurement, it would be impossible to exert any control over an operation on an ongoing basis (Slack et al., 2010). Some measures are used for the efficient strategic steering of a firm, while other measures are used for communicating the proper worth of a business to all interested parties (Askar et al., 2009).

All operations have an interest in keeping their costs as low as is compatible with the levels of quality, speed, dependability and flexibility that their customers require (Slack et al., 2010). Cost therefore appears as a performance indicator that can be used in manufacturing. Combined with its common-sized nature, cost is an ideal parameter for the measurement of the aggregate performance of the manufacturing sector.

Business performance measurement and control systems are the formal, information-based routines and procedures managers use to maintain or alter patterns in organizational activities. A typical performance measurement helps businesses in periodically setting business goals and then providing feedback to managers on progress towards those goals (Kellen, 2003). In the same vein, information is used at national level by various stakeholders in the crafting of strategies and policies that relate to industrial and economic development.

Uses Of Aggregate Performance Indicators

The following is a general discussion relating to the uses of various aggregate performance indicators. The uses outlined here relate to the indicators described above as well as others that have not specifically been mentioned here.

Labour related indicators are key determinants of living standards, measured as per capita income, and from this perspective are relevant to economic policy. These indicators reflect how efficiently labour is combined with other factors of production and how many of these other inputs are available per worker. (Organisation for Economic Co-operation and Development, 2001)

Multifactor indicators, for example capacity utilisation, help in demonstrating the contributions of labour, capital, intermediate inputs and technology to manufacturing. This helps in reviewing past performance and determining the potential for future performance (Organisation for Economic Co-operation and Development, 2001). This type of indicator is important in that it gives a guideline on the future outlook of the industry.

Among other things, aggregate performance indicators lead to an analysis of the increase or reduction in savings and investment that arises owing to productive activity, as well as the rise or fall in the capital-labour ratio, an important consideration in new investment in manufacturing (Organisation for Economic Co-operation and Development, 2001).

Aggregate performance indicators give an idea of the competitiveness of the sector and what factors need to be addressed to deal with enhancing competitiveness of the manufacturing sector (Confederation of Zimbabwe Industries, 2012).

In a broader context, performance indicators are used for political and policy reasons among which are lobbying of policy makers by business, providing Government with information on obstacles to economic growth as well possible solutions in designing better economic policy, and providing academia with information for analysing the current business environment (Confederation of Zimbabwe Industries, 2012).

As diversified as the users of aggregate performance data relating to manufacturing, the uses of the data are also well varied. In some cases, these data are not used in isolation but as part of a set of data, expanding their portfolio of uses, for instance in the development of policies and other economic documentation.

Users Of Aggregate Performance Indicators

While it is acknowledged that all people who are economically active will have an interest in the performance of the manufacturing sector, particularly for the reason that it is an important part of every supply chain in a modern economy, it is also the source of their sustenance.

Performance measurement metrics serve the different interests of different stakeholders. Some metrics are used as the basis for an organisational tune-up, quality improvement or business process reengineering (Askar et al., 2009). When considered in a broader context, on an aggregate level there are metrics that are used by shareholders, customers, vendors or creditors for evaluating the general quality of a provider or estimating the future growth of a firm (Askar et al., 2009).

In each case, the interests of the users vary but have commonality in that they relate to the future of the enterprise. In the case of the manufacturing sector in Zimbabwe, the same interest groups are keen to note the performance of the entire sector. In addition other stakeholders take an interest in these metrics. Of particular note, government departments, industry and trade bodies and foreign trade organisations are keenly interested in such metrics and make particular use of them.

Figure 2 .1 gives a graphic representation of some of the most significant users of aggregate performance indicators.

Figure 2.1: Users Of Aggregate Performance Indicators In Manufacturing

The list of users of aggregate performance indicators mentioned in Figure 2 .1 is not comprehensive but summarises the users based on the major uses to which these data can reasonably be expected to be put to.

In each case, aggregate performance indicators are used primarily for some sort of decision making. It is on the basis of this fact, and the assumption that the majority of users are interested in such data for more than just a mere appreciation, that an analysis of the accuracy and adequacy of the indictors is undertaken.

As input to decision making processes, aggregate performance indicators should satisfy at least minimum conditions of accuracy and adequacy that facilitate effective decision making. In this regard, the indicators should provide sufficiently detailed information to prevent users from suffering from indecision among other things. The following section describes decision making models and how the accuracy and adequacy of aggregate performance indicators is of material substance to decision making in the uses described above.

Decision Making

Aggregate performance indicators are a source of information for decision making. This is a very important source of information for organisations in the business of manufacturing as well as those maintaining relationships with the manufacturing sector.

As an alternative to aggregate performance indicators, manufacturers and other interested parties could gather information from various independent sources so as to have a clear impression of the prospects in manufacturing. This would invariably be a costly and time consuming effort and at some point the parties interested in such information would abandon the search for information having determined that the information they have been able to collect is sufficient for their purposes.

Knowledge creation and information management should be issues at the front of managers’ minds as potential sources of improving competitiveness (Johnson et al., 2008). Information is vital to the formulation and implementation of strategies and might impact on three important elements of a business’ operations:

ensuring that products/services are valued by customers,

outperforming competitors, and

making capabilities difficult to imitate (Johnson et al., 2008).

The importance of market knowledge to competitiveness applies in all sectors of industry, commerce and the public services (Johnson et al., 2008).

Given this background, accurate and complete information is essential for the development of strategies and policies relating to manufacturing, and aggregate performance indicators are the sources of some of this information.

Rational Decision Making Model

Human beings are rational within limits and cognitively limited, such that they cannot possibly be as comprehensive in information gathering and analysis as economists have long assumed. Human beings satisfice rather than optimise and search for information only to the point where they can find an acceptable solution. They often don’t keep on looking for the perfectly optimal solution to a problem they face. In many situations heuristics and rules of thumb are employed to make important decisions (Roberto, 2009).

For this reason, given the importance of the manufacturing sector, the production of accurate aggregate performance indicators is essential in ensuring that good decisions are made relating to manufacturing.

Adair (2010) proposes a five step process for making effective decisions. The five steps are:

Define the objective,

Collect relevant information,

Generate feasible options,

Make the decision, and

Implement and evaluate.

Aggregate performance indicators enter the decision making process in the second stage of the process during which relevant information is collected.

Regarding the collection of information, Adair (2010) states that not all information is immediately apparent. He distinguishes what is critically important information from available information, advising against the use of available information at the expense of relevant information.

In the case of aggregate performance indicators there is an apparent widespread use of available information at the expense of relevant information. John Robertson (Chiriga, 2011), with reference to capacity utilisation information published by the CZI, notes that the base which was being used to calculate the perceived industrial revival was wrong in the first place. This suggests that users of aggregate performance indicators were more often than not using available information as opposed to information which is relevant. The result is that decisions based on this information may be subject to failure.

In order to gather relevant information, one would have to do research, which will result in the decision maker incurring costs in time and money (Adair, 2010).

Policy and strategic decisions, particularly as they relate to something as important as the entire manufacturing sector of the economy, are made rationally, with a considerable amount of effort being applied to each step.

The description of the aggregate performance indicators given above suggest that a lot of relevant information is not provided by the indicators individually, and when used together, some of the indicators seem to give divergent perspectives which complicate the rational decision making process as much as the absence of any information at all.

For the purposes of decision making, the indicators described above may be deemed inadequate this conclusion is however verified and refuted later in the research from the findings of field research.

Intuitive Decision Making

An alternative approach to decision making is intuitive. This is based on the decision maker using their intuition in a highly unstructured manner to make decisions.

Given the implications of decisions relating to strategy and policy pertaining to the manufacturing sector, intuitive decision making is by no means ideal as a framework for decision making.

Chapter Summary

The foregoing review of literature has revealed inadequacies in the various metrics used to measure the aggregate performance of the manufacturing sector. The metrics, while good for the reason that they reveal an essential part of the performance of the manufacturing sector, do not give a holistic impression of the sector.

Comparisons of the metrics also reveal some inaccuracies that are inherent in the metrics.

These findings are confirmed or refuted by means of field research involving soliciting feedback from the various user groups and compilers of information which is collated into aggregate performance indicators.

CHAPTER THREE: METHODOLOGY

METHODOLOGY

Introduction

This chapter outlines the methods used to gather and analyse information used to verify or refute the findings made in the foregoing chapter.

The preceding chapter has highlighted the need for a combination of qualitative and quantitative methods to gather and analyses data in this research.

Research Philosophy

Two philosophical approaches have been used in the research:

Subjectivism, and

Positivism

Subjectivism

The subject matter considered in the research is of interest to several distinct user groups. Each of the user groups have different world views relating to the subject matter under consideration in this research giving them different perceptions of the subject matter. These different interpretations are likely to affect the actions of each of the user groups and the nature of their social interaction with others (Saunders et al., 2009).

The research and in particular the interpretation of findings from field studies conducted in the research is premised on the subjectivism philosophy owing to the fact that while the concepts of accuracy and adequacy which are considered in the research may have absolute definitions, their interpretation and application is subject to human judgement and the application of discretion on the part of each user group, giving the research an overtly subjective character.

Positivism

The research makes use of observed and observable data upon which inferences have been made. The results of the research are presented as scientifically determined generalisations. In addition, the findings of the research have been generalised into hypotheses the acceptance or rejection of which form the final relating to the population for which the study relates. These hypotheses lend themselves to the development of further theory relating to the subject of the research, and provide a basis for testing by means of further research.

Given the independence from and lack of interest in the subject matter of the researcher, the positivism philosophy is applicable to the present research. The philosophy is based on the assumption that the researcher is independent from and neither affects nor is affected by the subject of the research (Remenyi et al. 1998:33) cited in (Saunders et al., 2009).

The application of two research philosophies is consistent with the joint quantitative and qualitative nature of the research, wherein the positivist philosophy is the root of the quantitative element of the research and the qualitative element of the research stems from the subjectivist philosophy.

Research Design

The research is exploratory in nature. The research is explorative in so far as it explores an area that has not previously been researched extensively. The research diverts from being explanatory for the reason that it does not ultimately seek to draw conclusions on relationships between any particular variables. Notwithstanding the lack of relationship-explaining characteristics, the research makes use of explanatory concepts in determining the relationships between the various performance measurement indicators in an effort to determine the accuracy of the same performance measurement indicators.

An exploratory study is particularly useful in clarifying the understanding of a problem, such that the precise nature of the problem is understood (Saunders et al., 2009). In particular, the present research seeks to explore the precise nature of the accuracy and adequacy of aggregate performance indicators used to measure the performance of the manufacturing sector in Zimbabwe. In recent times the accuracy and adequacy of these metrics has come into question but the doubts have not been backed by research findings.

The research is of an applied nature in so far as its results are applicable to real life situations in the present day and the results of the research can find immediate application in the normal course of business.

Research Strategy

The survey strategy which is used in the present research is usually associated with the deductive approach, tending to be used for exploratory and descriptive research (Saunders et al., 2009). This strategy is consistent with the philosophy and design of the research which have been set out above.

A survey allows for economical collection of information from the relatively large research population which is described and enumerated in Table 3 .1. In the same way, the strategy is perceived as being authoritative owing to the fact that it ensures the inclusion of the views of a significant proportion of the population under consideration.

Using this strategy, quantitative data is collected and analysed using descriptive and inferential statistics to test the hypotheses relating to the research.

Data Collection Methods

Primary sources of data are used throughout the research. The collection of data has been conducted by means of interviews, focus groups and questionnaires; allowing for the collection of both qualitative and quantitative data.

Interviews

Face to face and telephone interviews were used to collect qualitative information from users who include policy makers, industrialist and economic and business consultants. These methods of data collection were most appropriate for the particular purpose owing to the fact that these particular groups from which data was being collected had perspectives relating to the subject matter that had previously not been taken into consideration and through interviews these perspectives would be revealed.

Owing to the cost and logistical challenges of setting up and conducting face to face interviews, telephone interview where also used to gather the required data. Where necessary, interviews were conducted over the internet using voice-over internet protocol applications.

The interviews used in this research were of a structured nature. A series of predetermined questions relating to the subject matter were posed to the members of the sample, who were requested to provide responses during the interview sessions. Provisions were made for respondents to contribute additional information that was relevant to the research that had not explicitly been solicited in the structured section of the interview.

In the case of face to face interviews, audio recordings of the interviews were be made so as to provide records of the interactions.

Focus Groups

Owing to the scant nature of information relating to aggregate performance measurement in manufacturing, focus groups that consisted of the collectors and users of such data were used to gather information relating to the adequacy and accuracy of such information.

Focus group discussions were setup on internet based discussion forums as these were easily accessible to members of the sample. The focus groups were setup on Linkedin, an internet based professional networking platform. The focus groups were run on the LinkedIn discussion forum in groups whose membership consists of Zimbabwean professional engaged in commerce and industry and with an interest in Zimbabwe’s manufacturing performance. Discussion logs were retained as evidence of the interactions

The information from the focus groups has been analysed qualitatively. Thematic areas were defined for the systematic analysis of the data collected from the focus groups and interviews.

Questionnaires

Questionnaires were used to collect information from the bulk of the members of the sample.

The Questionnaires included questions structured in the following ways:

Short response questions. These questions required the respondents to provide short responses to the questions posed. To allow for quantitative analysis of the responses, the responses require an affirmative or negative response as part of each response.

Likert scale styled questions. These questions were structured on the basis of five point Likert scales, requiring respondents to answer questions on the basis of their level of agreement with the questions posed.

Demographic questions. These questions required the respondents to provide details about themselves and the organisation that they represent, which were pertinent to the credibility and accuracy of their responses.

Questionnaires were distributed electronically as documents that could be downloaded by the respondents, completed and returned electronically.

Electronic distribution of questionnaires allowed a large sample to be used in the research without the research being expensive to administer. Electronically submitted completed questionnaires allowed for quicker collation and analysis of the responses.

Research Population

The research population consists of firms engaged in manufacturing operations, industrial and economic policy makers, industry representative organisation and business and economic consultants and other stakeholders. Table 3 .1 provides details of each of the populations.

Table 3.1: Research Population

Description

Population Size

Government Ministry / Department

5

Manufacturing Company

1060

Industry Representative Organisation / Lobbying Group

5

Academic Institution

20

Investment / Finance / Banking Institution

30

Private and non-governmental development organisations

20

Business / Economic Consultancy

50

The population is distributed around Zimbabwe, with large groupings of the various groups being concentrated in the various industrial centres around the country. The research included respondents from all the significant industrial centres in Zimbabwe; including but not limited to Harare, Bulawayo, Gweru, Kadoma, Mutare, and Chegutu.

Sampling

The research made use of a combination of probabilistic and non-probabilistic sampling techniques. With probability samples the chance, or probability, of each case being selected from the population is known and is usually equal for all cases (Saunders et al., 2009), whereas with non-probabilistic samples, the members of the population do not have equal chances of being selected. The sampling methods were used for the quantitative and qualitative elements of the research respectively.

Non-Probabilistic Sampling

Non-probabilistic sampling was used to determine the samples in the case of data intended for qualitative analysis. Purposive sampling was used in the present research. This enabled the use of judgement to select population elements that would best answer the research questions (Saunders et al., 2009). Deviant sampling has been used in this case on the premise that the data collected from the respondents will enable the collection of the most relevant data relating to the study, allowing for effective answering of the research questions.

The identification of the sample elements was done by assessment of the experience and continued involvement of the members of the population in manufacturing and policy development.

A sample size of at least 20% of the estimated population size in each category was used in the qualitative part of the research. This sample size was deemed sufficient as it ensured that at least twelve interviews were conducted from this homogenous group. Saunders et al. (2009) note that for research where the aim is to understand commonalities within a fairly homogenous group, twelve in-depth interviews should suffice.

Given that the population under consideration in this instance, industry representative and lobby groups, private and non-governmental industrial development organisations, relevant government ministries, and independent business and economic analysts, form a homogenous group in so far as they all use aggregate performance indicators relating to manufacturing in the same general manner; to lobby government in the areas of industrial and economic policies and for policy and strategy development, the recommended twelve interviews was deemed sufficient for the research.

The corresponding number of sample elements from each group is indicated in Table 3 .1.

Probability Sampling

The sampling frame for the quantitative element of the research consisted of manufacturing firms operating in Zimbabwe. The research excluded firms that were closed, or which were not producing any products at the time of the research.

The size of the sample frame was estimated at 1060 firms based on statistics presented by Mafunga (2009).

The sample frame was drawn from membership lists of the Confederation of Zimbabwe Industries, Zimbabwe National Chamber of Commerce and Association of Businesses in Zimbabwe, as well as lists of operational manufacturing companies supplied by the Ministry of Industry and Commerce.

A 95% confidence interval was desired with a 5% margin of error. It was anticipated that 75% of the respondents would respond in support of the alternative hypothesis, with 25% responding in favour of the null hypothesis. A response rate of 70% was expected. The corresponding sample size was given by the equation below:

…Equation 3.1

Where:

n – Minimum sample size,

p%– Anticipated positive feedback to null hypothesis = 25%

q% – Anticipated negative feedback to null hypothesis = 75%

z – The z value corresponding to the level of confidence required = 1.96

e% - the z value corresponding to the level of confidence required = 5%

The minimum sample size, n, was 288.12.

The adjusted minimum sample size was given by the equation below.

…Equation 3.2

Where

N - Total population = 1060.

The adjusted minimum sample size was 226.54.

The actual sample size used in the research was given by the equation below.

...Equation 3.3

Where:

na - Actual sample size required

n’ - Adjusted minimum sample size = 226.54

re% - Estimated response rate expressed as a percentage = 70%

The actual minimum sample size was 324.

Simple random sampling was used to select elements of the sample. This method of sampling ensured that each of the population elements had equal chance of being selected into the sample.

Table 3 .2 shows the sample sizes that were used for each of the populations.

Table 3.2: Sample Sizes

Description

Population Size

Government Ministry / Department

5

Manufacturing Company

1060

Industry Representative Organisation / Lobbying Group

5

Academic Institution

20

Investment / Finance / Banking Institution

30

Private and non-governmental development organisations

20

Business / Economic Consultancy

50

Data Analysis

The data collected by the methods outlined was been analysed using a combination of inductive and deductive tools, drawing conclusions on the entire population for which the sample relates.

Inductive tools have been used to analyse data of a qualitative nature, with deductive tools being used to analyse data of a qualitative nature. Below is an explanation of the rationale behind the use of either one of the approaches.

Inductive Analysis

Inductive analysis has been used so as to avoid overreliance on theoretical propositions in drawing conclusion on the subject matter. The inductive analysis allows for consideration of the social and economic context which the population exists in.

Deductive Analysis

Deductive analysis is used to explain the conceptual framework which is premised on the precept that aggregate performance indicators will reveal the same performance trends. In this case, the analysis was intended to determine if the pattern in the research data matches the expectation of the conceptual framework.

Reliability Testing

Prior to the analysis of the data collected by the methods described in section 2.5, an analysis of the reliability of the data was conducted. Reliability in this case was used to refer to the extent to which data collection techniques or analysis procedures yielded consistent findings (Saunders et al., 2009). In the present context it was assessed by posing questions relating to:

The ability of the measures to yield the same results on other occasions,

The ability of other observers to reach similar observations, and

The transparency in how sense is made from the raw data.

In order to ensure the reliability of the data, the following measures were taken:

Ensuring the anonymity of questionnaire respondents. The distribution of questionnaires discretely, directly to respondents and the collection of questionnaires directly from respondents ensured the anonymity of respondents. To assure respondents of such anonymity, a warranty of confidentiality was placed on the questionnaires, informing respondents that their responses would only be used for the particular research and would not be made available to third parties without their express consent.

Use of highly structured data collection methods. The tools used to collect data were of a highly structured nature, with exactly the same tools being used to collect data from respondents as per the distinctions of qualitative and quantitative data collection.

Qualitative Analysis

The qualitative data collected from the sample was analysed thematically. The thematic areas were determined from the responses given in interviews and focus group discussions.

The data was analysed on the strength of the experience of the respondents who put forward the data, as well as the evidence presented to corroborate the data.

Where conflicting perspectives were presented which were corroborated by equally valid and strong evidence, conclusions were drawn based on the frequency with which such views were expressed by different groups of respondents.

Quantitative Analysis

The quantitative data was analysed statistically and presented with the aid of visual tools that made the comprehension and assimilation of the data easier.

The demographic information received from respondents was cumulated and presented graphically to show the reliability of the responses. Normal distributions of demographics were used to determine the reliability of the responses. As such, each pertinent piece of demographic information was modelled as a normal distribution, following which an analysis of the responses was completed.

This was followed by a cumulative analysis of the affirmative and negative responses given to the short response questions. In the case of each question, a majority of responses served as the basis for drawing conclusions.

The Likert styled responses were analysed by means of Modal analysis, wherein conclusions were drawn on the strength of the most frequently given response to each question.

Following the Modal analysis, correlation analysis was used to determine relationships between subject matter that were related. This includes but is not limited to determination of the correlations between various metrics used to measure the performance of Zimbabwe’s manufacturing sector.

Hypothesis Testing

The Hypothesise, Test, Action, Business, HTAB system, was used to test the hypotheses stated in section Error: Reference source not found. Establishing the hypotheses encompassed all activities that lead up to the establishment of the statistical hypotheses being tested. Conducting the test involved the selection of the proper statistical test, setting the value of alpha, establishing a decision rule, gathering sample data, and computing the statistical analysis. Taking statistical action involved making statistical decisions about whether or not to reject the null hypothesis based on the outcome of the statistical test. Determining the business implications related to deciding what the statistical actions meant in business terms: interpreting the statistical outcome in terms of business decision making (Black, 2010).

Following this approach, the hypotheses were tested by means of modal analysis with acceptance and rejection of the hypotheses being based on a 95% confidence interval with a 5% margin of error.

The qualitative information served as a basis for corroborating or refuting the findings of the quantitative analysis. While the qualitative information was not used to test the hypotheses, the information was used to support the findings.

Figure 3 .2 is a summary of the methodology used in the present research, with branches representing the approaches used in each case.

Figure 3.2: Summary Of Research Methodology

Chapter Summary

The foregoing is a description of the methodologies used in structuring the field research from which results and the conclusions presented in the research have been drawn. The consistency of the methodologies presented above with the preliminary findings and requirements of the research structure presented in the review of literature is the basis of the initial validation of the research and the recommendations and conclusions drawn from them.

CHAPTER FOUR: DATA PRESENTATION, ANALYSIS AND INTERPRETATION

DATA PRESENTATION, ANALYSIS AND INTERPRETATION

Introduction

This chapter provides a summary, analysis and interpretation of the data collected by means of field research.

Demographics

The data relating to the surveys was analysed quantitatively to determine its consistency with the minimum sample size requirements, as well as to determine the desired findings of the surveys. Below is a summary of the demographic information relating to the sample used to come up with results and draw inferences relating to the population under consideration.

Table 4.3: Survey Response Rates

Description

Population Size

Sample size

Number of responses received

Government Ministry / Department

5

2

0

Manufacturing Company

1060

324

10

Industry Representative Organisation / Lobbying Group

5

2

0

Academic Institution

20

7

4

Investment / Finance / Banking Institution

30

10

0

Private and non-governmental development organisations

20

7

1

Business / Economic Consultancy

50

17

1

The response rates for the surveys, ranging from

3%

to

60%

are sufficiently high to constitute representative samples, allowing quantitative analyses within the parameters described in Section above.

The results of the survey are therefore accurate for a sample of the population within a 95% confidence interval with 5% margin of error.

The following sections give specific details of the constitution of the sample used in the survey, validating the survey results which are presented herein.

Length Of Time In Manufacturing

Responses to the survey were received from individuals who have been engaged in manufacturing or had maintained an interest in manufacturing for different lengths of time. A great number of respondents,

100.0%

, had been engaged in manufacturing for more than one year, during which time it is expected that they had some interaction with aggregate performance data and used such data.

In addition, a good number of the respondents,

52.9%

, had been involved with manufacturing for in excess of eight years, meaning that they were likely to have come across and possibly used aggregate performance data before and during the period of interest in this research, 2006 to 2012. Table 4 .4 provides details of the number of respondents, grouped by the length of time with which they have been involved with manufacturing in Zimbabwe.

Table 4.4: Respondents’ Length Of Time Engaged With The Manufacturing Sector

Length of time in manufacturing

Less than 1 year

1 - 5 years

6 - 8 years

More than 8 years

Number of respondents

0

7

1

9

Percentage of sample

0.0%

41.2%

5.9%

52.9%

Respondents who had been involved with manufacturing for less than one year were least likely to have an appreciation of the issues of greatest importance in the measurement of the sector’s performance. The proportion of this group of respondents was small and this means that the survey was not affected significantly by the inclusion of information which was not relatively informed by relevant experience of the industry.

Respondents who had been involved with manufacturing for between one and five years were likely to have an appreciation of the performance of the sector during the period under review, from 2006 to 2012. Such respondents may not necessarily have had an appreciation of the performance of the manufacturing sector prior to 2006, and would not be best suited to give comparisons of the performance measurement of the sector before and after 2006. This was a smaller proportion than that which has an appreciation of the performance of the manufacturing sector before and after 2006, having has six years or more continuous engagement with manufacturing.

Respondents with six to eight years involvement with the manufacturing sector were likely to have had an appreciation of the transitions that took place in manufacturing ushering in the current performance measurement regime. Respondents with more than eight years of involvement with the manufacturing sector were likely to have an appreciation of the performance of the manufacturing sector under different macro-economic conditions, allowing them to give informed opinions and comparisons on the measurement of the aggregate performance of the manufacturing sector before and after 2006.

Type Of Organisation

Responses were received from people a wide range of positions in different organisations. Table 4 .5 shows the relative proportions of respondents by the type of organisations they were in and represented at the time of the research.

Table 4.5: Type Of Organisation

Organisation Type

Government Ministry / Department

Manufacturing Company

Industry Representative Organisation / Lobbying Group

Academic Institution

Investment / Finance / Banking Institution

Retailer of locally manufactured products

Business / Economic Consultancy

Other

Number of respondents

0

10

0

4

0

1

1

2

Percentage of sample

0.0%

55.6%

0.0%

22.2%

0.0%

5.6%

5.6%

11.1%

The distribution of responses between the various user groups of aggregate performance data suggests that there was adequate representation of the opinions of the most critical users of such data. This serves to validate the sample used in the survey.

In addition to the distribution among different types of organisations, the distribution of respondents by the size of the organisations they represented at the time of the research was used to ensure that the opinions of smaller organisations is taken into consideration.

Size Of Organisation

Table 4 .6 shows the relative proportion of respondents by the size of the organisations they represented at the time of the research. The table shows fair representation of all organisation sizes. This had the effect of eliminating biases associated with data collected from organisations of a single size.

The research revealed that in the case of surveys such as the CZI’s Manufacturing Sector Survey, large firms were more likely to be surveyed than small and micro-sized firms and for this reason the present research was more representative than comparable studies.

Table 4.6: Size Of Organisation

Size of organisation

Micro Business

Small Business

Medium Sized Business

Large Business

Number of respondents

1

1

0

15

Percentage of respondents

5.9%

5.9%

0.0%

88.2%

Membership Of Representative And Lobbying Groups

A great proportion of the organisations represented in the survey were members of industry representative organisations and lobbying groups. By the nature of such groups, members were more likely to participate in surveys relating to aggregate performance than none-members. Given this background, the relative proportions of respondents whose organisations were members of various bodies compared to those whose organisations are not members of such bodies ensured a good balance of experience with the provision and use of aggregate performance data, and new perspectives regarding aggregate performance of manufacturing.

Table 4 .7 shows the relative proportions of organisational membership to industry representative bodies and lobbying groups that have an interest in the manufacturing sector.

Table 4.7: Membership Of Industry Representative Organisations

Organisations

ABUZ

CZI

ZNCC

None

Other

Number of references

1

5

3

4

6

Percentage of memberships

5.3%

26.3%

15.8%

21.1%

31.6%

It should be noted that any given organisation can be a member of more than one body. As such, Table 4 .7 represents relative proportions as opposed to the absolute numbers of respondents who participated in the study. Table 4 .7 indicates which of the listed organisations is most influential.

While one or more representative group may evidently have greater membership among members of the sample, suggesting greater influence throughout the population under study, these bodies may have agendas that compromise their objectivity and therefore may not necessarily be better sources of information than those bodies with considerably smaller membership.

It should be noted that industry representative organisations and lobbying groups exist for the purpose of representing members’ interests; at times doing so may be at the expense of other stakeholders.

The proportion of respondents that were not members of any representative and lobby groups served to represent opinions that were not necessarily congruent with those of industry representative organisations and lobbying groups. The proportion of respondents that were not members of any representative or lobbying groups gave sufficient polarisation of the research findings to validate the objectivity of the results.

Survey Participation

A significant proportion of the respondents’ organisations had participated in surveys to determine the aggregate performance of the manufacturing sector in the eight years prior to the research. Of the respondents, a small proportion had also personally supplied data in such surveys. Table 4 .8 shows the relative proportion of organisations that had participated in surveys intended to determine the aggregate performance of the manufacturing sector. Table 4 .9 shows the ratio of respondents who had personally supplied information in such surveys. Table 4 .8 and Table 4 .9 showed that the present research was sufficiently representative of organisations that had and had not participated in surveys in the eight year period preceding the present research, as well as the proportion of respondents who had and had not supplied information in such surveys thus validating the survey.

Organisations and individuals who have supplied information in a survey were likely to have had a different, possibly better, appreciation of the meaning, appropriate use of, and relevance of the information than those who have not. In addition, such organisations and individuals were likely to have a greater appreciation of the shortcomings of the surveys.

56.3%

of the respondents indicated that the organisations that collated data relating to the aggregate performance of the manufacturing sector did not provide adequate explanations to organisations and individuals requested to provide information regarding the surveys. Table 4 .10 shows the proportions of respondents who indicated that the collating organisations provided sufficient explanation and those who indicated the otherwise. The table also includes the proportion of respondents who were not sure about the sufficiency of explanation of the information requirements. The proportion that was not sure was attributed to organisations and individuals that had never supplied information for surveys, nor had been privy to the processes used in collecting such information.

Table 4.8: Survey Participation Statistics

Survey

CZI Manufacturing Sector Survey

Zimstat surveys

Ministry of Industry and Commerce surveys

None

Other

Number of participants

6

5

9

3

1

Percentage of participation

25.0%

20.8%

37.5%

12.5%

4.2%

Table 4.9: Supply Of Information For Surveys

Response

Yes

No

Number of respondents

7

10

Percentage of respondents

41.2%

58.8%

Table 4.10: Adequate Instruction Given By Compiling Organisations

Response

Yes

No

Not Sure

Number of respondents

6

9

1

Percentage of responses

37.5%

56.3%

6.3%

Uses Of Aggregate Performance Data

Respondents indicated the uses to which they put aggregate performance data. The data evidently was used for a plethora of purposes, with the most frequently referred to being strategic management. Table 4 .11 shows the relative proportions of uses of aggregate performance data by the respondents. The data in the table indicates that while there are a number of uses, those of a strategic or policy nature are the most significant uses of this data, accounting for

50.0%

of the total usage of aggregate performance data.

This statistics suggests that the data should be more of a strategic than operational nature. As such, the data should be more inclined towards global indicators such as financial indicators, as opposed to localised factors such as resource utilisations.

The data reveal that the main uses of aggregate performance data are strategic management, comparative analysis, and policy development. The data also reveal that a considerable proportion of the population are interested in aggregate performance data relating to the manufacturing sector solely for the purpose of getting a general appreciation of the performance of the sector.

Table 4.11: Uses Of Aggregate Performance Data

Use

General appreciation

Strategic management

Policy development and review

Lobbying

Reporting to stakeholders

Comparative analysis

Production planning

Other

Number of references

7

10

5

5

4

7

2

0

Percentage of uses

17.5%

25.0%

12.5%

12.5%

10.0%

17.5%

5.0%

0.0%

Communication Of Aggregate Performance Data

Respondents confirmed receiving information about the aggregate performance of the manufacturing sector through a number of official and unofficial media. The source from which information was received had implications on the accuracy of the information, with some sources being prone to distort or misrepresent information.

Table 4 .12 indicated the proportions of respondents that receive information through the various media that are available.

Table 4.12: Communication Of Aggregate Performance Data

Medium of communication

Official publications

Direct Communication

Consultative meetings

Through industry representative groups

Internet

Hearsay

Other

Number of references

6

4

2

2

4

3

0

Percentage of references

28.6%

19.0%

9.5%

9.5%

19.0%

14.3%

0.0%

57.1%

of respondents received information through official channels created by the organisations that were responsible for the collation of aggregate performance data. This meant that this proportion of respondents was less likely to receive distorted or misrepresented information. A further

9.5%

of respondents received information from industry representative bodies and lobby groups. While information from these sources was likely to be distorted in favour is the interests of the respective groups, it is noted that the degree of such distortion would not be so great as to discredit the data.

A small proportion of respondents received communication of aggregate performance from other sources which were not always known to be reliable, and which were prone to distortion owing to them being easily tampered with seeing as they were in the public domain.

33.3%

of the respondents indicated that they receive communication about aggregate performance data from internet sources and hearsay, as well as other sources which do not constitute official channels. These sources are least likely to present aggregate performance data in its entirety. In the case of newspapers for example, the publication of aggregate performance data is at times for the purposes of drawing attention to particular issues that are of contemporary interest. This type of presentation of information invariably leads to the omission of some data, changing the context in which the data should be used. As such, the data becomes distorted and its use compromised.

In addition to the mere communication of aggregate performance data, respondents overwhelmingly reported that the organisations that collated aggregate performance data did not adequately explain how the data should be used and its limitations. Table 4 .13 shows the proportion of respondents that indicated that the organisations that collated aggregate performance data explained the uses and limitation of such data adequately to those that expressed that the uses and limitations of such data was not adequately explained. The overwhelming finding that the data’s uses and limitations were not adequately explained led to the conclusion that the data was prone to abuse and no effort has been made to ensure the proper use of such data.

While aggregate performance data should have universal applicability, the nature of the data makes it suitable for some applications and not others and the uses should therefore be categorised such that different users exercise caution in the aggregate performance data.

Table 4.13: Explanation Of Aggregate Performance Data And Its Limitations

Uses and Limitations Explained

Yes

No

Number of responses

4

12

Percentage of responses

25.0%

75.0%

Frequency Of Use Of Aggregate Performance Data

62.5%

of respondents indicated that they used aggregate performance data more frequently than on an annual basis this suggests that such data should be made available more regularly than annually. However,

6.3%

indicating that they never use such data, with the remaining

37.5%

indicating that they refer to the data annually. Table 4 .14 shows the proportions of users who referred to aggregate performance data during various time intervals.

Given this situation, various metrics, particularly GDP, Capacity Utilisation and Unemployment, were evidently inadequate for the purposes of the majority of the users owing to the fact that these data are updated annually at best. For this reason it stood to reason that these data are not frequently referred to by the majority of members of the population.

Table 4.14: Frequency Of Reference To Aggregate Performance Data

Frequency

Never

Monthly

Quarterly

Annually

Number of responses

1

2

7

6

Percentage of responses

6.3%

12.5%

43.8%

37.5%

Having determined the viability of the data collected in the surveys, the following sections give an analysis of the data.

The sections that follow detail the analysis conducted on the survey data, providing conclusive results which have been generalised for the entire population at a 95% level of confidence with a 5% margin of error.

Quantitative Analysis

In the surveys, ten questions were used to determine the sentiments of the sample elements. Eight of the questions were structured on the five point Likert scales, with the other two being structured as short response questions. Responses from the eight Likert styled questions were subjected to statistical analyses, serving as the basis for determining the quantitative results of the study. The two short-response questions were analyses qualitatively.

Having used Likert styled questions, modal analysis was used.

Statistics: Modal Analysis

A number of statistical methods were used to generate the results from the survey data. Table 4 .15 gives a summary of the responses from the survey. Details of the questions to which the data in Table 4 .15 relate are provided in the appendix in section Error: Reference source not found.

Table 4.15: Summary Of Responses

Question Number

Total responses In Agreement

Total responses Undecided

Total Responses Not in Agreement

Most Frequently Appearing Response

12

11

0

6

Agree

13

6

5

6

Agree

14

5

8

4

Neutral

15

8

6

3

Agree

16

6

8

3

Neutral

17

7

7

3

Neutral

18

6

5

6

Disagree

19

4

5

8

Disagree

Conclusions have been drawn on the basis of the modal response to each question, with the statistics of total numbers in agreement and not in agreement with the statement in the question, being used to affirm or refute the findings of the analysis given by modal analysis.

In cases where a majority of responses were neutral, qualitative analysis was used to determine the definitive sentiment of the population.

Hypothesis Test

In Section Error: Reference source not found, two hypotheses were proffered relating to the accuracy and adequacy of aggregate performance indicators used in measuring the performance of Zimbabwe’s manufacturing sector.

The following is an analysis of the data collected though survey to determine which of the hypotheses holds true.

The hypotheses being tested are:

H0 – The aggregate performance indicators used to measure the performance of the manufacturing sector in Zimbabwe are accurate and adequate.

Alternatively, it was hypothesised that:

H1 – The aggregate performance indicators used to measure the performance of the manufacturing sector in Zimbabwe are inaccurate and inadequate.

The basis of accepting or refuting the hypotheses was an absolute majority of responses affirming or disputing the accuracy and adequacy of the performance indicators. An absolute majority would be constituted of at least 51% of the sample.

Accuracy and adequacy of aggregate performance indicators have been considered separately owing to the fact that they are mutually exclusive concepts. As such, the absolute majority in terms of both accuracy and adequacy is given by the product of an absolute majority of both accuracy and adequacy.

As such, the hypotheses were restated mathematically as:

H0: p ≥ 0.51*0.51, and

H1: p< 0.51*0.51.

Where:

p – Proportion of the sample that believes that the aggregate performance indicators used to measure the performance of the manufacturing sector in Zimbabwe were accurate and adequate. This proportion is given on the scale from zero, 0.00, to one, 1.00.

A single-tail test was used to test the hypotheses.

A test alpha value of 0.05 was used in the hypothesis test. This alpha value was consistent with the confidence interval and margin of error that has been used in determining the sample size in the survey.

The critical value used to designate the rejection region was:

Z = 1.96

This Z values were extracted from statistical tables relating to the standard normal distribution. Figure 4 .3 shows the acceptance and rejection regions for the null hypothesis.

Figure 4.3: Hypothesis Acceptance And Rejection Regions

The test statistic value, Zcalc, for the survey findings was given by the equation below.

…Equation 4.4

Where:

Zcalc – Test Statistic

p – Sample proportion or proportion of desired outcome

q – Proportion of undesired outcome

n – Sample size

π – Hypothesised population proportion

The proportion of desired outcome for the sample and the proportion of the undesired outcome are given in Table 4 .16.

Table 4.16: Summary Of Accuracy And Adequacy Of Indicators

Indicator

Accurate and Adequate

Inaccurate and Inadequate

Capacity Utilisation

0.152

0.166

Gross Domestic Product

0.069

0.111

Employment Statistics

0.083

0.166

All aggregate performance indicators

0.093

0.104

The sample statistics, Zcalc, for the population under consideration are given in Table 4 .17.

Table 4.17: Sample Statistic And Null Hypothesis Decision For Various Metrics

Metric

Capacity Utilisation

GDP

Employment Statistics

All aggregate performance indicators

Zcalc

-1.24

-3.10

-2.65

-2.36

Null Hypothesis Decision

Reject

Reject

Reject

Reject

Given that Zcalc> 1.96, the result falls outside the acceptance region. As such, the null hypothesis, H0, is refuted and the alternative hypothesis, H1, is accepted.

The findings of the quantitative analysis were corroborated by a qualitative research which confirmed that the metrics used to measure the performance of the manufacturing sector in Zimbabwe are both inaccurate and inadequate.

The following sections provide details of findings of the qualitative component of the research.

Qualitative Analysis

The data used for qualitative analysis was collected from individuals who had considerable experience working in the manufacturing sector in Zimbabwe. Data was collected by means of interviews which were conducted telephonically, in-vivo, as well as through VOIP applications, focus group discussions which were setup on Linkedin.

The data collected related specifically to the type of data that should be used for determining the aggregate performance of the manufacturing sector, the current regime of measuring such performance, as well as alternative sources of information that were being used to supplement aggregate performance data.

The qualitative data collected revealed that the indicators used for the measurement of the performance of the manufacturing sector in Zimbabwe were both inaccurate and inadequate for their intended purposes.

Respondents in the study proffered numerous reasons for discrediting the accuracy of the indicators currently in use, referring particularly to disparities that exist between reported performance of the manufacturing sector and the actual performance of the sector. The data presented by respondents corroborated the facts presented in the literature review which made use of comparisons of the finding of different performance indicators.

The qualitative data collected provided a differ approach for the analysis of the accuracy and adequacy of the performance indicators, complementing the findings of the quantitative research and verifying the facts presented in the literature review. While the qualitative component of the study did not involve comprehensive numerical or statistical analysis, the fact that the finding of this study were consistent with those of a scientific approach served to prove beyond any doubt that the indicators in question lacked viability in terms of their accuracy as well as their adequacy.

The data collected in the qualitative part of the study was generally related to one of several distinct areas of interest. The data was therefore classed into the respective interest areas and analysed. The thematic areas under which data was analysed were:

operational efficiency,

comparative analysis,

international trade,

policy, and

technology.

The following is a summary of the findings relating to each of the thematic areas.

Operational Efficiency

The research identified the need to take into consideration issues of operational efficiency when measuring the aggregate performance of the manufacturing sector. In this regard, factors such as labour productivity, labour turnover, energy efficiency, energy and water consumption, and the reliability of plant and machinery.

The research drew attention to the fact that one of the most important measures of the performance of the manufacturing sector was the cost of production and this is a function of the efficiency of the production processes. Taking into consideration the fact that the reliability of plant and equipment deteriorates with age, and a considerable amount of the equipment that is currently being run by manufacturing firms across the country is old, operational efficiency is a major consideration when measuring the performance of the Zimbabwean manufacturing sector.

While the plant and equipment being run in the manufacturing sector is not th