THE DEVELOPMENT OF A MANAGEMENT ERROR ORIENTATION QUESTIONNAIRE

The aim of this study was to develop a Management Error Orientation Questionnaire (MEOQ) to be used as an instrument to measure the attitude of management towards errors in the workplace. The sample comprised of 232 managers from a variety of business sectors. A factor analysis on 59 items yielded three factors and these factors were interpreted as the attitude of dealing with errors, the risk of errors and error strain. The three scales were subjected to an item analysis and yielded significant levels of reliability. Opsomming Die doel van die studie was om ’n Bestuursfout Orientasie Vraelys (Management Error Orientation Questionniare) te ontwikkel wat as ’n meetinstrument kan dien om die houding van bestuur met betrekking tot foute, in die werksomgewing te bepaal. Die steekproef het uit 232 bestuurders uit ’n verskeidenheid van besigheidsektore bestaan. ’n Faktoranalise op 59 items het drie faktore opgelewer, naamlik die houding aangaande die hantering van foute, die risiko van foute en foutspanning. Die drie faktore is onderwerp aan ’n itemanalise en het aanvaarbare vlakke van betroubaarheid gelewer.

The importance of attitudes (or its allied concepts such as world view, orientation and others) in the study of errors can be inferred from the present-day approach to management or organisational behaviour where values and culture are considered as primary factors in determining organisational success and survival.On the basis of the literature there is the temptation to suggest that value and attitude are concepts applicable to the individual and that value and culture are related concepts applicable to the organisation.
Errors have been defined as "… all those occasions in which a planned sequence of mental or physical activities fails to achieve its intended outcome, and when these cannot be attributed to the intervention of some change agency" (Reason, 1990, p. 9).Rasmussen (1986, p. 149) defined error in a much simpler way: "… if a system performs less satisfactory than it normally doesbecause of human act or a disturbance that could have been counteracted by a reasonable human act -the cause will very likely be identified as a human error".

Nature of errors
According to Reason (1990) human error is neither as abundant nor as varied as its vast potential might suggest.Not only are errors much scarcer than correct actions, they also tend to take a surprisingly limited number of forms.Errors also appear in similar guises across a wide range of mental activities.Comparable error forms can be identified in action, speech, perception, recognition, judgement, problem solving, decisionmaking, concept formation, interpretation and the like (Rasmussen, 1982).
The exact representation of a typical framework for human error is not possible.Human error and human cognition is related and a typical description of the aspects of human cognition is not yet fully understood.Notwithstanding the differences between psychologists on these subjects regarding structure, processing and representation, Reason (1988Reason ( , 1990) ) and Rasmussen (1986) indicated that most of the contemporary models contain important areas of common ground, which could be classified as control modes, cognitive structures, attentional mode and schematic control mode.
The accuracy of error prediction depends largely on the extent to which the factors giving rise to the errors are understood.This understanding requires a theory, which relates to the three major elements in the production of an error: (a) the nature of the task and its environmental circumstances, (b) the mechanisms governing performance and (c) the general nature of the individual (Reason, 1990).
The notion of intention and error is inseparable and intention comprises of two elements: (a) an expression of the end-state to be obtained, and (b) an indication of the means by which it is to be achieved.Both elements may vary widely in their degree of specificity (Reason, 1990)."All intentional actions have intentions in action but not all intentional actions have prior intentions" (Searle as cited by Reason, 1990, p. 6).A distinction between error types and error forms is also used to classify errors.Error type relates to the presumed origin of the error within the stages involved in conceiving and then carrying out an action sequence.The stages are described by Reason (1990) under three broad headings: planning, storage, and execution.

Error theories
The literature abounds to a number of taxonomies reflecting a variety of practical concerns and theoretical orientations and ranging from the highly task specific to broad statements of underlying error tendencies.The grouping of error theories could be done by means of the natural science tradition or the cognitive science tradition.The basic assumptions about human cognition form the bases of the modern error theories (Reason, 1990).
A number of models were developed in the theory on errors (Neisser,1976;Broadbent, 1984;Norman and Draper, 1986).The most practical model was the one developed by Rasmussen (1982) who developed the skill-ruleknowledge framework in an effort to construct an executable taxonomy of human errors.The framework defined three performance levels: Skill-based level (SB).At this level human performance is governed by stored patterns of pre-programmed instructions.
Errors at this level will be related to the intrinsic variability of force, space or time coordination.
Rule-based level (RB).Solutions of familiar problems are governed by stored rules.Errors at this level are associated with the misclassification of situations leading to the application of the wrong rule or with the incorrect recall of procedures.

Knowledge based level (KB).
Actions must be planned on-line, using conscious analytical processes and previous gained knowledge or experience.Errors would therefore arise from resource limitations or incomplete or incorrect knowledge.Rasmussen (1982) identified eight stages of decision-making (or problem solution): activation, observation, identification, interpretation, evaluation, goal selection, procedure selection and activation.The major contribution from this framework has been to chart the shortcuts that human decision makers take in real-life situations.

Evaluation of errors
The focus on the evaluation of errors is stressed for two main reasons.First, errors have been deeply investigated from different theoretical perspectives and in a number of environmental settings (Reason, 1990).It is thus possible to refer to data already collected and to consolidated theories to interpret and even to classify a given deviation from the correct course of action.Second, errors are worthwhile indicators signalling all those circumstances in which environmental characteristics do not match with cognitive abilities (Rizzo, Parlangeli, Marchigiani, & Bagnara, 1996).
With few exceptions, error analyses address mainly the quantitative issue (Norman, 1983).These analyses cope with the number or the rate of errors that are produced in a given session of interaction, but leave to the research teams in charge for finding a solution for the different types of errors.
The few approaches that have been devised to address the qualitative issue mainly focus on the occurrence of the error, not on its detection and recovery.They provide guidelines for reducing the occurrence of errors or for minimizing their impact but rarely for supporting their detection and recovery (Rizzo, et al., 1996).
One of the few set of guidelines that cope with the issue of error detection is the one put forward by Norman (1983).This set of principles has, however, some drawbacks that make the principles only partially applicable to all the types of human error.Indeed, the guidelines are applicable to 'slips' and rarely to 'mistakes' (Rizzo, et al., 1996).

Error orientation
Errors and attitudes are important issues in work psychology for a number of reasons.Errors produce stress, lead to accidents, influences quality outputs, has an effect on performance and will affect the culture of the organisation (van Dyck, 1997).
The attitudes towards errors and how one deal with them are indications of a company's organisational culture (Baron, 1986).Bureaucratic companies usually have the attitude to prevent errors from happening at all costs, while the entrepreneurial business usually have a more positive attitude towards error and what can be learnt from them (van Dyck, Frese and Sonnentag, 2000).If a company attempts to change its culture one needs a measure of error orientation (Van Dyck, 2000).Rybowiak, Garst, Frese and Batinic (1999) Rybowiak et al., (1999) developed one of the first questionnaires aimed at measuring the orientation towards errors -the Error Orientation Questionnaire (EOQ).The publication indicates that error orientation can be conceptualised within the general coping concept.According to this research team, the coping concept refers to the extent that an individual can stay calm in the face of errors, to cover up the fact that an error has occurred, as opposed to communicating about them, and to actively deal with an error or to learn from it.
In the development of the EOQ (Rybowiak, et al., 1999) a set of six scales -error competence, learning from error, error risk taking, error strain, error anticipation, covering up errors -for measuring orientation towards errors.From the items included in the scales, they suggest that error competence and learning from errors are related to self-efficacy, plan-and action orientation; error risk taking was related to need for achievement, control rejection and readiness for change; error strain was related to ill-health; error anticipation implied a negative outlook on life; and covering up was related both to environmental factors (career stress, job security) and to lack of self-esteem, rejecting responsibility and control.
According to the authors the strength of their study was the fact that they could cross-validate the factor structure and that the scales were developed in a representative sample.
Previous studies have indicated that an error culture does exist in companies (Rybowiak, Garst, Frese and Batinic 1999;Van Dyck, 2000).Van Dyck (2000) developed an Error Culture Questionnaire and concluded that beliefs, attitudes, norms and behavioural approaches regarding errors could be shared and that error culture does exist.The reaction of management toward errors could be placed in two basic categoriespunishment and empathy.
The replication and the function of exploratory behaviour, and the effect of training were researched.Dormann & Frese, (1994) and Van Dyck, (2000) published results on the managing of error culture in organisations.Bear (1999) studied the influence of error management climate and psychological safety climate on the relationship between modern manufacturing practices and company performance.Other studies include error training and the role of goal orientation (Heimbeck, 1999), and the role of error culture on organisational performance (Van Dyck, et al., 2000).

Attitudes
Attitudes are a central part of human individuality.In most instances it is most certainly the case that a persons' attitude towards a particular attitude object may influence his or her behaviour towards this object.Moreover an attitude toward one object may influence behaviour (and attitudes) towards other attitude objects (Oppenheim, 1992).Attitudes are central to the subject of social psychology because they are central to social lives.Individuals are easily categorised according to their attitudes (e.g.conservatives or feminists).
An attitudes statement is "… a single sentence that expresses a point of view, a belief, a preference, a judgement, an emotional feeling, or a view on something" (Oppenheim, 1992, p. 11).Bohner and Wänke (2002) simply define an attitude as being a summary evaluation of an object of thought.Baron (1986, p. 140) defines attitudes as "… relatively lasting clusters of feelings, beliefs, and behaviour tendencies directed toward specific persons, ideas, objects or groups." Attitudes could be developed through natural selection (Tesser, 1993) or through external influences which include exposure (Bornstein, 1989), contiguity, imitation, reinforcement, observation (Bohner & Wänke, 2002) and persuasion (Howard, 1997).

Attitude and behaviour
Various indicators of attitude strength as well as personality variables have been identified as moderators of the attitudebehaviour relation (Bohner & Wänke, 2002).High correlations though between attitude and behaviour do not provide sufficient evidence for inferring that attitude cause behaviour.
Behaviour may influence attitude.Third variables such as salient context-dependent beliefs influence both attitude reports and behaviour.Schwartz and Bohner (2001) indicate that if these context factors remain stable, high attitude-behaviour correlations could be expected.Attitudes do predict behaviour but they do not cause behaviour.
Attitudes are mostly perceived as straight lines, running from positive, through neutral, to negative feelings about the object or issue in question.The general attempts at measurements are then to place a person's attitude on the linear continuum in such a way that it can be described as mildly positive, strongly negative and so on -preferably in terms of a numerical score or else by ranking (Oppenheim, 1992).
Attitudes will play an important role in the operations and management of a business.Work-related attitudes often play a key role in shaping the behaviour and actions in organisations.
The understanding of attitudes of individuals will enhance the understanding of the people side of the business.The views individuals hold about their jobs and organisations often exert powerful effects upon their performance and other aspects of organisational behaviour.The most successful methods to change attitudes are based on approaches of persuasion and dissonance (Baron, 1986).
Key work-related attitudes that indicate the role attitudes can play in the business and management environment includes job satisfaction, organisational commitment, prejudice and dissonance.
On the basis of the importance attached to the role of attitudes, values and culture as far as errors are concerned, the main purpose of the study then was to develop a questionnaire through which the attitudes of managers towards errors can be measured.
The study addresses the issue of the attitudes of management towards errors and a questionnaire will be developed to measure this attitude.

METHOD Participants
The sample was chosen from a variety of business sectors in South Africa.A basic database with potential companies was drawn up.No distinction was made in terms of the size of the company and the selection of businesses included large multinational corporations, public companies, private companies and individually owned businesses.
Managers at the level of middle management, senior management and at executive level were approached within the identified businesses to complete the questionnaire.It was expected that the respondent has some subordinates reporting to him/her or at least be in decision-making managerial position.

Procedure
A senior executive of the target companies was contacted and after receiving confirmation that the company could be approached for the research, the questionnaire was distributed through electronic mail to either a contact person in the company (who would further distribute it within the company) or to the contact list supplied by the company.The questionnaire was returned to a dedicated electronic mail address.In some cases hard copies of the questionnaire were distributed when requested.
It is difficult to determine an accurate response rate for electronically distributed questionnaires since the number of questionnaires forwarded from one participant to the next, can not be established.The researcher sent out 1386 questionnaires electronically to 283 businesses and 150 hard copies were distributed to individuals.In total 247 questionnaires were returned of which 85% were returned electronically.Of the 247 questionnaires that were returned, fifteen were discarded due to the incorrect completion thereof and 232 were statistically analysed.The composition of the population and sample is presented in Table 1.The sector analysis indicate that the manufacturing environment dominated the sample as 25% of the participants came from this sector and none of the other sectors reaching more than 10% participation.From Table 1 it can be seen that the majority of the participants are general managers from the senior management level and that they have a tertiary qualification, are male and in the age group 36 to 45.

The Measuring Instrument
The questionnaire developed in this study consists of two sections -a biographical section and the Management Error Orientation Questionnaire (MEOQ).The Management Error Orientation Questionnaire (MEOQ) builds on research of individual error management orientation (Rybowiak et al., 1999).
The MEOQ consists of 59 items, which were developed within eleven theoretical attitude factors related directly to the management of errors.These factors are error competence, learning from errors, error risk taking, error strain, error anticipation, covering up errors, error communication, thinking about errors, helping with errors, prevention of errors and general attitude towards error.
The first six of these categories were taken from the Error Orientation Questionnaire (EOQ) that was developed by Rybowiak et al. (1999).Each category has a minimum of four questions.
All items in the MEOQ are endorsed on a five point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Statistical analysis
The factor analysis on item scores was conducted by the Statistical Consultation Service of the Rand Afrikaans University according to a procedure suggested by Schepers (1992).

Factor analysis of items
The 59 items of the MEOQ were intercorrelated and subjected to a principal factor analysis.Due to a lack of space, the intercorrelation matrix is not reproduced here.
Based on Kaiser's criterion (1961), 13 factors were postulated and extracted.The obtained factor matrix was rotated to simple structure by means of a Varimax rotation.
Subscores were calculated in respect of 12 factors with acceptable loadings.Five items were omitted during the initial factor analysis.
Factor analysis of the subscores of the MEOQ The 12 subscores were intercorrelated and subjected to a principal factor analysis.The obtained factor matrix was rotated to simple structure by means of a Direct Oblimin rotation.The intercorrelations of the subscores are given in Table 2.
Three factors were postulated and extracted.The three factors accounted for 51,97% of the variance of the test scores.The eigenvalues are presented in Table 3.
Table 3 indicates that three eigenvalues are greater than unity.Table 4 shows the rotated factor matrix of the MEOQ.From an inspection of Table 4 it appears that all three factors are relatively well determined.Factor I has six moderate to high loadings and Factor II and III have three loadings each.The intercorrelations between the three factors are given in Table 5.Two items were omitted during the second-order factor analysis.Table 5 shows that Factor II (the risk of errors) and Factor III (error strain) is positively correlated.
Three scales were formed corresponding to the three factors by including all the items with high loadings on each factor.An item analysis was performed in respect of Scale I with 37 items (Table 6), Scale II with 6 items (Table 7) and Scale III with 9 items (Table 8) and Cronbach alpha coefficients of 0,937; 0,701 and 0,717 respectively were found.
Table 6 shows that in Scale I the item-total correlations range from 0,118 to 0,812.The scale is internally relatively inconsistent and no items were rejected.The high reliability coefficient was sufficient for the purpose of constructing a MEOQ.,54,30,45,28,51,12,14,8,27,57,46,53,38,59,3,11,7,SFS2 49,4,56,39,36  In Scale II the item-total correlations range from 0,275 to 0,622 and although internally consistent and no items were rejected, there were only six items in this scale and the reliability coefficient was just above the standard of 0,7.The scale does fit the purpose of constructing the MEOQ and was retained.The obtained reliability for Scale III was 0,717.The item-total correlations varied from 0,224 to 0,464, which indicate internal consistency.No items were lost and the scale comprises nine items.
MANAGEMENT ERROR ORIENTATION QUESTIONNAIRE 41

DISCUSSION
The 59 items written for the MEOQ were subjected to a factor analysis and the results obtained in the present study yielded three factors.The three factors that emerged were identified as Factor I: The Attitude of dealing with errors, Factor II: The Risk of Errors and Factor III: Error Strain (stressed caused by errors).
A content analysis of the theoretical categories indicate that more items would have been lost (26 against 7) if the original theoretical design would have been used as the MEOQ.In the exploratory factor analysis one of the final factors that was formed (Factor III), was named Error Stain and in the content analysis the factor error strain with 9 items were eliminated.
The Attitude of dealing with errors can be said to represent aspects like competency in dealing with errors, the communication of errors, the prevention of errors, thinking about errors, helping with errors and general attitude about errors.This factor relates to the practical aspects of error management.
The Risk of Errors is an indication of the effect that errors could have on the business and is specifically related to management.This factor addresses risk management and The results of this study cannot be compared to other similar research.Rybowiak et al. (1999) developed the EOQ with which the orientation of individuals in terms of errors was tested.Their study indicated six constructs in the empirical model of the EOQ and the results of the present study supported the reliabilities that were found in relation of the EOQ.Although differently worded, the three scales that were found to be reliable in the MEOQ also correlated well in the EOQ.It was found in the MEOQ that the cross-validation of the eight empirical groupings that were originally used in EOQ, was not reliable and that only two of those theoretical factors of the EOQ would have correlated well with the final factors of the MEOQ if they were to be used in the MEOQ.As the three scales found in the MEOQ were done through exploratory factor analysis, and the final questionnaire included most of the questions that were used in the EOQ, it is concluded that the exploratory scales of the MEOQ carries more weight than the scales in the EOQ.It was also concluded that error orientation differs between individuals (as was tested in the EOQ) and managers (as was tested in the MEOQ) and that different factors determine the attitude between individuals and managers.
In practical terms the MEOQ could be used when new managers are appointed, in career contingency plans, in the development of a management team and when individuals are being promoted in the corporate hierarchy.The instrument could be used in the strategic planning process of a company to determine level of error orientation as an important internal factor and could indicate whether it is a strong point or weak point within the organisation.
In conclusion, the present study opens up new research possibilities, for instance the determination of management orientation among gender groups and in the South African context among different race groups.The testing of management attitudes in high-risk business sectors should further be determined.

TABLE 6 ITEM
STATISTICS OF SCALE I: THE ATTITUDE OF DEALING WITH ERORS On the surface it appears that there is a connection between Risk of Errors and Error Strain with a positive correlation between these factors.The Attitude of dealing with errors will be preventative in nature and the Risk of Error and Error Strain relates to attitudes once the error has happened.

TABLE 7 FACTOR
SCALE II (STRESS CAUSED BY ERROR): DESCRIPTIVE STATISTICS

TABLE 8 FACTOR
SCALE III (THE RISK OF MAKING ERRORS): DESCRIPTIVE STATISTICS