The symptoms of and consequences to selection errors in recruitment decisions

The ability to select effective human capital has long been viewed as the bedrock of the human resource function and a driver of organisational success. There is a plethora of literature available on recruitment and selection methodologies but little empirical evidence of the consequences to selection errors. Data was gathered on 393 incidents of selection errors across a wide range of industries. The symptoms and attribution of, and the wide-ranging consequences to, these selection errors were documented. Methods to rectify the selection error and the outcome of those attempts were also explored. Recommendations based on these empirical findings are offered.


Introduction
One of key features of the knowledge era is the increased mobility of knowledge workers particularly amongst those who have rare skills and competencies, which leads to higher levels of recruitment and selection. Recruitment and selection are some of the most critical human resources decisions an organisation can make (Sutherland & Jordaan, 2004). Recruitment is the process of attracting and encouraging potential employees to apply for a position, while selection is the process of making fair and relevant assessments of the strengths and weaknesses of applicants with the intention to hire them (Boxall & Purcell 2008;Breaugh & Starke 2000). Correct selection creates a match between the capabilities and inclinations of prospective candidates against the demands and rewards inherent in the organisation. The effective appointment of appropriate candidates is critical to organisational success yet few organisations formally evaluate their recruitment and selection efforts (Carlson, Connerly & Mecham, 2002;Hacker, 1997). According to Dale (2003), appointment decisions are amongst the most important a manager has to make as they affect the ability to achieve organisational targets, the quality of services or products delivered to customers and the well being of the staff. Huselid (1995) showed a relationship between successful hiring and financial performance. Some organisations calculate the costs of appointments using metrics such as selection ratios and responses to adverts, but the outcomes of wrong selection decisions are not typically assessed and empirical research in this area is scant.
Most managers have experienced the problems that result from selection errors but often continue making them. "Most companies are so determined to prove that their hiring system is foolproof that they not only fail to admit to hiring mistakes, they also keep them around longer than they should in the vain hope that they may yet work out" (Buchen, 2007:80). Selection errors often occur because of a mismatch between expectations and reality for both parties (Blenkinsopp & Zdunczyk, 2005). Despite huge amounts of time, effort and costs allocated to the recruitment and the selection process, recruitment practices in many firms produce an unacceptable failure rate that reduces business competitiveness and corporate profits (Davis, 2005;Ryan & Tippins, 2004;Boxall & Purcell, 2008). Cole, Field, Giles and Harris (2009) find that recruiters infer dispositional characteristics from a candidate's resume and use these inferences to judge a candidate's employability. However, these inferences are found to be largely invalid and unreliable. The best recruitment and selection techniques have a validity coefficient of 0,6 (Dale, 2003) so even when selection processes have been carefully designed and executed, hiring errors can occur.
The ability of the manager making the decision is a key component of the selection decision and there is a significant amount of research that shows that this is the most common area where selection mistakes occur. The manager then lives with the consequences of a wrong selection decision. Accurate evidence about human resource practices is a requirement for understanding what is taking place and making correct interpretations (Ryan & Tippins, 2004). The movement towards evidence based management means translating learnings from best or worst practice into organisational principles (Rynes, Giluk, & Brown, 2007). This research set out to document empirically the causes, symptoms and consequences to selection errors from the perspective of the manager. The aim being that the evidence may assist human resource practitioners, line managers and academics to understand and appropriately respond to and possibly reduce the occurrence of selection errors.

Recruitment sources
In recruitment, choices are made between a number of sources, such as: internal promotions, hiring via leads given by current staff, getting human resource departments to handle the recruitment process, or using external recruitment agencies and head hunters. The method of recruitment can lead to differences in the success of recruiting efforts as they all have their advantages and disadvantages (Sims, 2002;Russo, Gorter, Nijkamp & Rietveld, 1997). Moser (2005) and Chan (2006), compare the outcomes of internal and external candidates. They found different levels of unmet expectations between those who came via internal versus external channels. The above authors say the usage of recruitment sources can have various impacts on both pre and post hire outcomes e.g. met expectations, job satisfaction organisational commitment, job performance and employment survival with employees who had more internal information about the organisation being better prepared for transition into the organisation.

Selection techniques
The selection process is a key source of expectations, many of which may be inaccurate (Blenkinsopp & Zdunczyk, 2005). Cable, Aiman-Smith, Mulvey and Edwards (2000) discuss how organisations often lead applicants to believe favourable rather than accurate beliefs about the organisational culture in order to get them to join the organisation. This in itself may lead to misfits between candidates and the firm. Bossidy (2001: 48) says "I feel strongly that the interview is the most flawed process in American business". He says that the success rate of executive hire is at most 70%. Ones and Dilchert (2009) concur with the failure rate of selection processes and discuss the predictive ability of different selection techniques.
Selection processes come in many guises but the fundamental issue is how to make selection more reliable, how to define performance appropriately and how to use techniques that improve the firm's ability to predict which individuals will be good performers (Boxall & Purcell, 2008). Despite the large amount of empirical research in selection approaches (Robertson & Smith, 2001), Ryan and Tippins (2004), show that techniques with a high predictive value are seldom adopted by human resource practitioners and managers and that contextual factors such as budget, time constraints, employee demographic demands and legal requirements play a role in the likelihood of adoption of best practices. They discuss how the various stakeholders in the selection decision, such as hiring managers, applicants, legal departments and unions exert differing influences on the decision. Lockyer and Scholarios (2007), Piotrowski and Armstrong (2006), König, Klehe, Berchtold and Kleinmann (2010), explore a range of findings around best practice in selection, the scientist-practitioner gap in the use of selection procedures and find that organisations choose selection methods based on likely applicant reactions, costs and diffusion of methods. Predictive validity, organisational self-promotion and perceived legality were found to be less important.

Person-job and person-organisation fit approaches
The ideal employee is defined as one who "fits" with his work environment (Chaung & Sackett, 2005;Goodman & Svyantek, 1999). The term "person-environment fit" (PE fit) has been studied extensively, particularly the components of Person-Job fit (P-J) and Person-Organisation fit (P-O). P-O fit is the compatibility between people and organisations (Kristof, 1996). Carless (2005) links the attraction, selection and retention of individuals to similarity between the person and their work environment. This includes phenomena such as personality, attributes and values of the individual and organisational values, goals structures processes and culture. P-J fit is the match between individual knowledge, skills and abilities and demands of the job (Carless, 2005). Chuang and Sackett (2005), find that P-J fit is perceived as more important than P-O in the early stages of recruitment, but that this changes as the recruitment process continues, with P-O gaining in importance during later stages. Researchers Chaung and Sackett (2005), Kristof-Brown (2000), Carless (2005), Hoffman and Woehr (2006) have studied the relationship between P-O and P-J and job satisfaction, turnover, stress, organisational commitment, intention to quit. Kristof-Brown, Zimmerman and Johnson (2005), show that there is a strong correlation between job satisfaction and organisational commitment when there is a high P-J fit and an inverse relationship between P-J fit and intention to quit and that P-O fit correlates well with job satisfaction and organisational commitment but has a moderate correlation with intention to quit. Our study looks at whether P-J and P-O fit may be involved in the selection errors.

Consequences of poor selection
There are three aspects to the consequences of selection errors:

Attribution of the error
Attribution theory is concerned with the way we explain or account for outcomes and causality as being either internal or external to oneself (Moerdyk & Mashinini, 2002). Failure is often attributed to external factors such as the behaviour or internal attributes of others rather than to our own behaviour. Manzoni and Barsoux (1998) argue that managers are often complicit in the employees' lack of success and that when an employee fails or performs poorly; managers typically assume the problem to be the employee's fault and responsibility. This perspective may play a role in the identification of, response to and the possibility of learning from, a selection error. Bossidy (2001) says it is uncommon to take time to learn from hiring errors and he cautions that companies need to learn from their mistakes so previous selection errors are not repeated. Capelli (2002) discusses the need for process improvement in hiring by evaluating every step, pinpointing weaknesses, seeking their root causes and identifying opportunities for improvement. Dale (2003) suggests that various stakeholders, including recruitment agencies, HR departments, the manager and the employee, in the recruitment and selection process may share the blame for selection errors. No empirical literature could be found on the attribution of selection errors. Carlson et al. (2002), state that few organisations calculate the costs of selection errors. The emotional costs paid by the subordinate and the organisational cost associated with the failure can be long term and indirect (Manzoni & Barsoux, 1998). It is very difficult to put an exact value on all the potentially important positive consequences of making a good selection decision and the negatives of making a poor one. Jackson and Schuler (2003), find that a poor hiring decision can cost as much as five times the employee's salary. The US Department of Labour more conservatively estimates that a bad hiring decision equals 30 per cent of the employee's first years earning potential (Hacker, 1997). The more senior or more specialised the position the higher the costs are likely to be. Some of the negative outcomes of selection errors mentioned by Jackson and Schuler (2003); Hacker (1997) and Werther and Davis (1989) are: the employee performing poorly which leads to lost productivity, absenteeism, loss of self esteem by the employee, poor morale amongst peer workers as they suffer the consequences of the colleagues poor performance, customers expectations not being met, managers experience increased pressure and failing to meet their objectives, injuries and accidents may occur, possible lawsuits, even union activity and subsequent labour turnover leading to future recruitment costs. There is however scant literature giving empirical evidence of the consequences to selection errors.

Remedial actions to correct the selection error
Once a selection error is identified, (the time taken for this to occur can vary widely), managers need to decide on how to respond to it. There is a range of intervention possibilities with each option being an investment in time, with associated opportunity costs, resources and other associated costs such as legal suits or severance payments. Bossidy (2001), Davis (2005) and Hacker (1997) suggest that when a hiring mistake has been made corrective action needs to be taken swiftly to correct the hiring error. Dale (2003) suggests that methods to rectify a poor selection decision include developmental feedback, job redesign, redeployment, dismissal or termination by mutual agreement. The payback on the intervention will depend on its quality and key contextual factors. Possible outcomes are: improved performance, either significantly or marginally or when differences are irreconcilable, termination. However, finding new replacements may be costly and recurrent. Manzoni and Barsoux (1998), point out that if underperformance is not addressed there will be continued underperformance and resultant tensions, and that the consequences can have long term impacts for both parties. Organisations need to deal with negative consequences associated with psychological contract violations resulting from selection errors, as these have wide ranging consequences in the workplace (Blenkinsopp & Zdunczyk, 2005). No empirical research could be found on the efficacy of methods used to address selection errors.

Aim of the research
As shown above there is a plethora of studies on recruitment and selection practices but no empirical studies on selection errors could be found. Lievens and Chapman (in Wilkinson, Bacon, Redman & Snell, 2010) in their meta-analysis of recent recruitment and selection research literature discuss the macro and micro validity of recruitment and selection processes, but never mention the outcome of selection errors. Our research is focussed on increasing the understanding of a number of aspects of selection errors from empirical evidence. These are; some possible causes of selection errors, the time it takes to realise an error had occurred, the consequences of the errors, how they were responded to and what the outcome of these responses were, and the attribution of the error to the candidate or the manager who made the selection. Without attributing the error correctly it is unlikely that the manager or organisation will learn from the mistake and is likely to repeat it. Much of the recruitment research has been in single organisations or industries, making generalisations difficult. Our study deals with multiple industries and examines the decision, the recognition of the error and the remedy of it from the managers' point of view.

Methodology
The methodology used was a combination of the critical incident technique and a survey. The critical incident technique is a procedure, which facilitates the investigation of significant occurrences (events, incidents, processes or issues) identified by the respondent, the way they are managed, and the outcomes of the perceived effects (Chell, 1998). The objective is to gain an understanding of the incident, taking into account the cognitive, affective and behavioural elements. It is an exploratory inductive technique. The critical incident for this study was defined as a selection error having been made.
The population for the survey was identified as any line manager who felt they had made a significant selection error in the last year. The sample was obtained by judgemental sampling via South African MBA students, employed at over 80 companies in a wide range of industries, who interviewed managers who admitted to making a selection error (Zikmund, 2007). This ensured that respondents were from a wide range of companies and deals with the problems of an in-company survey, such as the impact of organisational culture, leadership and performance management systems on the process and outcomes of poor selection. The unit of analysis was the respondent managers' perceptions of the critical incidentthe selection error.
As recommended Babbie & Mouton (2004) prior to the development of the questionnaire, six in-depth interviews with senior HR managers and line managers were conducted to establish a valid set of constructs for the symptoms of selection errors and the remedies thereof. The survey questionnaire was designed based on these constructs and the literature review. The questionnaire was then pre-tested on five line managers for clarity and adequacy of response categories. The 5-page questionnaire consisted of closed ended questions, Likert scales and two opened ended questions. Three hundred and ninety three useable questionnaires were obtained.
Data was analysed using NCSS. Two variable chi-squared tests of dependence, using an alpha level of 5%, were run to look for relationships between variables (Zikmund, 2007). Content analysis was carried out on responses to the open ended questions to analyse the data in an objective, systematic way in order to identify and quantify recurring patterns and themes in the data (Zikmund, 2007).

Limitations
As the data was in the form of perceptions and self report, particularly in the area of how the selection error was responded to, it can be expected that some respondent bias may have been present in giving normative (Zikmund, 2007), self serving or perceived to be desirable answers. We caution on representatively as the sample was drawn from one large geographic area of South Africa, which is however the financial and industrial hub of the country. We only sourced the managers' perceptions and not those of the employees who were perceived to have been selection errors.

Discussion of results and findings
The 393 managers who were interviewed represent a wide spread of demographics. Eighteen per cent were executives, 39% general managers, 41% middle managers and 2% supervisors. Thirty per cent had been in management for more than 10 years, 33% between 6 and 10 years, 32% between 2 and 5 years and 6 per cent for less than two years. Eleven per cent were less than thirty years old, 68% were between the ages of 31 and 45, and 22% were older than 45. In terms of their experience in conducting selection interviews, 30% conduct less than 4 a year, 46% between 4 and 12 a year and 28% conduct more than 12 a year. Forty six per cent had received specific training in selection processes. The balance had not. In terms of size of the organisations represented, 30% had less than 100 staff, 24% between 100 and 500 staff and 46% more than 500 staff.
The details on the employees who were perceived to have been selection errors are; 7% were executives or senior managers, 27% were middle managers 22% were supervisors, and 45% were classified as being either technical or specialists. Sixty nine per cent of them had at least one degree. Forty six per cent of them were less than 30 years old, 47% between 31 and 45 years of age and 7% were older than 45. Table 1 indicates the range of recruitment methods used to select the employees deemed to be selection errors. All tables are in percentages.

Recruitment and selection methodologies
The manager's considerations, when selecting the employees classified as selection errors, in terms of the fit between applicant and both the organisation's culture and the job to be performed (Kristof-Brown et al., 2005) are given below.  By adding the data in the first two columns in the Table  above these results are somewhat startling in that 44.7% and 53% of respondents said that during the selection process they were hardly concerned whether the individual could do the job or would fit into the culture respectively. It must be remembered that this data is based on critical incident processes and hence these were all incidents of selection errors. Part of the cause of the hiring errors may be related to these findings of the scant attention being paid to P-O and P-J fit during the recruitment and selection process. The literature above indicates how important these variables are to a wide variety of organisational outcomes (Kristof-Brown et al., 2005).
Chi-squared testing for a relationship between whether the respondent had received training in selection processes and whether they were experienced in selecting staff and the above two variables showed there were no significant relationships i.e. these errors were made irrespective of their levels of interview training and experience. Ryan and Tippins (2004) and Dale (2003), state that best practice in human resources is seldom implemented. It is not that common for managers to be trained in all aspects of recruitment and selection such as P-O and P-J fit. Table 3 shows the time taken for the respondents to realise that a selection error had occurred. The majority (81%) of problems became evident in the first six months. Fitz-enz and Davison (2002), in their seminal work on measurement in human resource management suggest that the assessment of quality of a new employee can only begin 6 months after appointment. For our respondents, this was not the case. Selection errors were detected much earlier.

Evidence of selection errors
The two variable chi squared tests showed these time lines were unrelated to size of organisation and seniority of the appointee. This is an interesting finding as one could have assumed that it would take a longer time to discover a selection error the more senior one is in an organisation as consequence of error and span of control are used in many job evaluation systems to determine seniority (Grobler, Warnich, Carrell, Elbert & Hatfield, 2006).
Respondents were asked what the evidence was that a selection error had occurred. The question asked the respondent to choose any number of options from a range of possibilities. Other, please specify 13,6 The respondents ticked an average of 5.82 options showing firstly how widespread the symptoms of selection errors are and secondly how interrelated these variables are (Manzoni & Barsoux, 1998;Werther & Davis 1989). Some variables may be seen as inputs and outcomes to other variables. For example, not having the correct skills could lead to poor performance, which then becomes a source of discontent amongst colleagues.
The "other" answers were analysed via content analysis and the evidence reported fell into 5 categories: Firstly, unacceptable behaviour e.g. absenteeism, alcoholism, excessive sick leave and use of internet, harassment, alcoholism, theft, lying; religious evangelism to customers. Secondly, a variety of reasons indicating that the employees were disappointed with the nature of the work, the level of the work, reward -including feelings that the company owed them something or they were being undervalued. Thirdly, the employees displayed a lack of drive, passion, commitment and the ability to cope with pressure. Fourthly, the appointee not being a team player, and finally, a lack of leadership skills. This data again gives evidence for the literature on importance of and outcomes to selection decisions.

Responses to the selection error
The respondents were asked how they had responded to the selection error. They were asked to tick however many methods they had used from a list of options. There is likely to be some bias in this self-reporting. The results are shown in the table below. The respondents ticked on average 5.2 of the items indicating that they had tried many ways to handle the selection error. This supports the Dale (2003), statement that there are many options as to how to respond to selection errors.
The respondents were asked how the problem had been resolved. The outcome is shown below. The data shows that in 80 per cent of the cases the selection error had not been resolved in a manner that lead to performance being improved. This supports the work of Carlson et al. (2002), as well as Manzoni and Barsoux (1998), as discussed above. The costs of the termination of employment of 62.7% of selection errors would be considerable including lost productivity, costs associated with poor customer service, management time, training costs and lower morale amongst fellow employees, vacancy costs to the costs and subsequent recruitment costs (Sutherland & Jordaan, 2004). In many cases this labour turnover would represent a hiatus in the business unit if one measured the time from the start of the initial recruitment and selection process, to the appointment, subsequent poor performance, remediation attempts, exit and then re-recruiting and selecting another new appointee. Intangible costs include brand damage and impact on the workplace. As discussed above it is extremely hard to quantify these costs.
Chi squared testing with the outcome of the selection error Table 6 and the source of the appointee shown in Table 1, showed a significant relationship in that those promoted from within are disproportionately still in employment but are equally divided between those whose performance improved and those who are still performing poorly. This supports the view of Chan (2006), that internal appointees enjoy protection and insulation from external competition. The results however also show that 45% of those selected via promotions have left the company. This shows that promoting from within is not necessarily an ideal selection method and that the costs to both parties and the organisation's brand as an employee may be affected.
Chi-squared testing was done on the relationship between Tables 5 and 6 to explore the outcome of the various response behaviours. We found that intensive tailored training was the only method that was significantly related to the employee staying and their performance improving. The techniques associated with the employee leaving are mentoring and coaching, verbal and written warnings, dismissals, telling the employee they had no future with the company, and reaching financial settlements. Isolating or marginalizing the employee had the outcome of the employee staying in the job without an improvement in performance. Redesigning the employee's job did not have the intended consequences and was significantly related to either the employee leaving or staying and performing poorly. The other techniques were not related to the outcomes. What is clear is that intensive evidence based performance management techniques need to be inculcated into all line managers.

Impact of the selection error
The respondents were asked the extent to which the selection error had impacted on the organisation. Sixty three per cent of the respondents reported that the negative impact was minimal or felt only in the immediate work unit but in 37% of the cases the negative impact of a selection error was widely felt. There may be some under reporting in this data. Chi squared testing showed that the level of impact is unrelated to the time taken to realise there is a problem and the level of seniority of the appointee. The latter goes against what was expected because of the consequence of error factor which is commonly used in job evaluation systems (Grobler et al., 2006).
The respondents were then asked for more specific details of the impact of the selection errors. The results are shown in the rank ordered Table 8 below. The modal categories are shaded.  , customer dissatisfaction (59%) and staff morale is impacted by the selection errors. Increased training time and expenditure was reported by 53% and 50% mentioned damage to the organisation's image. Although no attempt was made to put exact costs on these ramifications they would impact the organisation broadly. These findings give empirical support to the writings of Carlson et al. (2002); Manzoni and Barsoux (1998) and Sutherland and Jordaan (2004).
The respondents' answers in this table show that they do not see the causal link firstly between customer dissatisfaction (second highest) and loss of revenue (second lowest) and secondly between increased costs of doing business (third lowest) with the input costs of increased management time, training cost increases and lower employee morale. This lack of understanding the causal linkages would decrease the likelihood of the line mangers being willing to learn from their mistakes and apply evidence based management techniques (Bossidy, 2001;Cappelli, 2002;Rynes et al., 2007).
Chi squared tests were done between this data and the organisational level of the employee. These only revealed two significant relationships; if the appointee is at the executive or senior manager level more "other employees" tend to leave the organisation and there is a greater loss of revenue.

Attribution of the selection error
The respondents were asked in an open-ended question to what they attributed the poor performance of the employee they viewed as a selection error. Seventy two per cent of the respondents completed this section. When content analysis was applied to the completed answers, it was found that only 9% of the managers felt they were responsible for the selection error while 91% felt that the candidate was to blame. This clearly demonstrates the validity of attribution theory, which states that when people succeed they attribute this to internal factors such as their ability and effort but when they fail they attribute it to factors external to themselves (Moerdyk & Mashinini, 2002;Martinko, Harvey & Douglas, 2007). This is also referred to as a self-serving bias (Robbins & Judge, 2007). Based on these theories it could be expected that the manager would be positive towards the new employee in the early stages of the relationship as they made the selection decision. This may change overtime as the manager observes poor performance and the cause of the misfit or poor performance then becomes attributed to the appointee.
Of the 25 managers who took some responsibility for the selection error, 23 blamed the problem on poor selection i.e. poor reference checking, poor interviewing, nepotism in selection, quota filling, lack of psychometric testing, people being promoted too soon and inexperience in selection. Other common explanations for the poor appointment were that the organisation doesn't adapt to newcomers easily in that the organisational culture is too strong or insensitive, the expectations of the organisation and/or employee were not spelled out sufficiently, and that the manager did not give sufficient feedback, training, mentoring and coaching.
The 91% of the responses blaming the appointee can be grouped into the following themes:  Misleading in selection process: 26 responses related that the appointees lied on their CV or in the interview.


Interpersonal problems: There were 11 responses referring to interpersonal problems and or social problems.
The categorisation of explanations for the poor performance relate to a great extent to a misfit between P-O and P-J fit. This supports the work of Chaung and Sackett (2005), Kristof-Brown (2000), Carless (2005) and Hoffman and Woehr (2006). As a high percentage of the respondents said they did not take these factors into account (Table 2) when making the selection decision this may reveal one of the causes of the selection errors. Figure 1 draws together the empirical findings of this study based on a large sample in many industries. It shows the linkages from the data above that result in a vicious cycle that begins with a selection error. The cycle shows how selection errors lead eventually to decreased organisational competitiveness via increasing costs and decreasing revenues which arise from the factors identified in the study. It shows the efficacy of various remediation attempts and asks the key question as to whether the organisation learns from the evidence of the consequences to the selection error in order to break the cycle and not repeat the selection error once again.

Figure 1: The vicious cycle of selection errors
This research has provided empirical evidence that selection errors are costly and difficult to remedy yet the number of errors remains high. The costs of and consequences to selection errors are considerably higher and wider ranging than previously shown in the literature. The findings above also reinforces Ryan and Tippins (2004), finding that managers do not implement best practice in recruitment and selection as well as not taking full responsibility for their role in making the selection.
We found that appointment errors end in a termination and/or continued poor performance 80% of the time. The data showed that all recruitment methods can lead to selection errors but that the lack of investigating P-O and P-J fit (Kristof-Brown et al., 2005) may be the cause of many of the problems. The errors made in the incidents may have been resolved by using a wider range of section devices. Ryan and Tippins (2004) and König et al., (2010) who found that despite the existence of high-validity selection methods, managers do not adopt them when making selection decisions. An additional complication is the evidence of attribution errors and managers seemingly blaming the appointee for the poor performance rather than themselves. This has consequences for the manager's readiness to learn how to avoid selection errors in the future (Bossidy 2001;Cappelli, 2002).
Our findings suggest the following recommendations for managers:


The consequences of selection errors need to be made known to managers and human resource professionals in order to convince all parties to adopt best practice in selection. Organisations need to continuously review the effectiveness of their selection methods and feed the results back to managers, either in the form of training or adding superior approaches to existing practice.
 A clear understanding of P-O and P-J dimensions and how these can be assessed in potential appointees would reduce the likelihood of selection errors.
 Selection errors can be detected in less than six months. The response to them needs to be rapid and well formulated based on evidence of what works best (Davis, 2005).
In conclusion much work has been done on the predictive value of recruitment and selection methods but little attention has been paid to the consequences of selection errors. We have shown that the problem with selecting incorrectly does not end with the appointment of the candidate and that if the firm does not act to recognise and remedy the error, the consequences are wide-spread. Future research into selection errors could focus on longitudinal studies on methods for managers to remedy and reduce the impact of selection errors as well as on the perceptions of employees who were deemed to be selection errors.