This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
ln recent years, selection process has become one of the most crucial parts of any organizational decision . Selection processes should only be based on the ability to make a
contribution to the organization's effectiveness and potential for development.'(Cornelius, 1999)
Every organization has different methods of carrying out selection and the choice of technique is largely dependent on the job role. Redman et al suggest that 'Best-Practice employee selection is usually associated with the psychometric model designed to accurately measure candidates '(2009).
Google uses a complex mix of psychometric test to judge the attitude, skill and achievements of an individual to make sure they match the job requirements. The author proposes to evaluate the selection process at Google inc. where the candidates are faced with a selection processes designed not by the HR team but, by a team of highly qualified engineers and computer specialist.
The company puts great emphasis on finding candidates with great achievements in their
life. The researcher aims to evaluate the selection process and conclude whether or not the process at Google is a valid and reliable process and suggest recommendations if any.
2. Aims and Objectives
The aim of the research is to evaluate the employee selection process at Google and identifying potential areas of further research and development.
The Objectives of the research in order to achieve the desired aim are :
Defining Selection as a Process
What is selection and what are the different selection techniques?
What influences a company's decision in choice of the selection method?
What is reliability and Validity of a selection process
2. Analysis of the current selection technique at Google
What is the current process of selection at Google.
Identifying the steps involved in the process.
Reason behind the choice of selection process.
3. Discussing the pro's and con's of the selection process at Google.
Positive aspect of the selection process at Google.
Analysing any possible flaws and assessing future need.
3. Literature Review
'Management in organizations is making decisions, and these may be fundamental element, or underlying building block of strategy making' (Leopold et al, 2009) and selection forms an important part of the entire decision making process. The main aim behind the selection is to get the right people in the right jobs and to retain them in the company. Mistakes do happen when attempting to select the most suited for the job, and hence it gets even more important to use the right method to selection. Some even argue the even before we choose the right selection method we need to attract the right candidate. The process of attracting right candidate is called as recruitment. Recruitment is described 'as the stage where the role is sold to the candidate and it is either lost or gained.'(Roberts, 1997. Dale, 2003).In the part 3.1.1 of the project the author highlights the importance of a good recruitment campaign as a strong base for selection.
3.1.1 Recruitment: first stage to Selection
Recruitment is the very first stage to facilitate good selection. According to Oxford dictionary recruitment is 'The action of finding new people to join an organization or support a cause' (Oxford Dictionary).
Many people get confused with recruitment and selection being the same thing. They are a part of each other, i.e. they are in fact two sides of a coin. Recruitment is the method of establishing contact between the applicant and the company, whereas selection is picking the most suited out of the applicants to do the job. In words of Leopold et al 'Recruitment activity filters applications through factors such as whether a realistic preview of the job is communicated, the ease or difficulty of application , the time scale for applicant response and the quality of the recruitment pack information'( Leopold et al, 2009)
The organizations need to identify many important points before proceeding with the recruitment process. The key question to answer in any recruitment activity is 'who do you want to attract'. Who exactly would you want for that job role, with what set of qualifications, experience and skills. Methods like Job descriptions, job analysis and person specifications help in identifying the basic requirements for the job role. All three terms are very similar to each other especially by the means of purpose which they aim to establish that is a clear picture of what is required by the job role in terms of the ideal candidate. According to Margaret Dale the recent job descriptions are' redefined to become the profile of the largest audience or segment of labour market.'(Dale,1995). 'Job analysis is not expected to enhance reliability, but a job analysis should enhance the amount of job information brought into the interview, thus decreasing deficiency. Similarly, by focusing the interview on job-related content, it should reduce contamination' (Campion et al, 1997)
The second question the organization needs to identify whether the wish to recruit internally or externally. Both the processes have their own advantages and disadvantages. Recruiting internally is always a cost effective way of recruiting as no money is spent advertising the vacancy to external sources and also the person chosen will be available to start work immediately, hence the gap is filled sooner. There are also strong arguments put against internal recruitment being 'lack of fresh blood coming in the organization, and managing situations where internal candidates are unsuccessful' (Taylor, 2005) . On the other hand external recruitment offers various methods that a company can choose from and the choice of method largely depends on' the audience targeted, time available to fill the position and finally money available for the advertising' (Taylor, 2005). Most modern and big companies use recruitment websites and professional link sites for recruitment. 'While companies have long turned to sites like Monster and CareerBuilder to fill open jobs, they're not always the best options for start-ups-especially those in specialized fields or that require specific skill sets'(Wang, 2011). Gareth Roberts also highlights the same concept explaining that 'recruiting in house or externally is simply a matter of judgment which needs to be taken into account, as in any other decision, of cost, convenience and suitability' (Roberts, 1997).
A recent trend in recruitment has been observed and as Julie Beardwell and Tim Claydon (2007) define it as '1.Ethnocentric (filling positions by the nationals of the parent company), 2.Polycentric (where host country nationals fill most position), 3.Regiocentric (decisions made on regional basis), 4.Geocentric (where best people are recruited irrespective of nationality)'. These patterns can be observed in many companies where the local population gets a preference over other applicants which may be due to the government regulations and one of the most common reasons observed especially in the United Kingdom is the visa restrictions that people migrating from other countries have and hence they are not offered jobs . But largely speaking these are issues that are many a times beyond control of the organizations themselves.
As mentioned earlier the companies use multiple advertising sources to source candidates for any one position. 'Despite the economy's impact on employment it's critical that recruiters continuously work hard to attract and compete for top performers' (Aurthur ,1998). They have a wide choice starting from newspapers to internet websites and from doing it via internal HR to hiring Job agencies. It would be very difficult to class one method as the idle for all companies and job roles. According to a survey done by Hays recruitment for CIPD in 2012 it has been found that ' While the effectiveness of methods to attract applicants varies according to organisation sector and size, the most effective methods overall were reported to be through organizations' own corporate websites and recruitment agencies, as was the case last year'(www.CIPD.co.uk). According to Adam Burroughs(2012) 'one must be very careful in choosing agency as a means of sourcing, and must pay attention to culture, size, location, industries served and services provided are factors to consider (right fit)'
Recruitment as highlighted before also acts as an elimination activity as it helps removing candidates that are not suitable for the position by creating job descriptions which are precise and show a clear aim the hope to achieve. But, the author argues that there is a possibility of the recruitment process being flawed and the company rather than gaining may lose the right candidate. Recruitment is a first stage and a process of flow of information for both the recruiter and the employee. It is this flow of information that needs to be controlled and shared with a lot of careful thought. When a candidate applies for a job and reads the information provided about the role and the company creates and image in the mind of the applicant for the company and this is the point where the decision of applying or not to has been made. According to Margaret dale(1997) research and experience has indicated that if clear messages are not transmitted early, the damage done by the mismatch can be costly to both the parties. Thus, as Torrington et al (1995)comment' A good recruitment system is crucial to the organization, not least because the recruitment of suitable employees will improve retention and morale among the existing workforce by accentuating to those inside and outside the organization the importance it attaches to people'.
3.1.2 Reliability and Validity of Selection
All employers are searching for the very best employees available, but very few know exactly how to increase their odds of finding just the right person (Mills, 2007). The 'right fit' as the modern HR uses to address the best candidate out of the total applications received. 'The crucial importance of selecting people who can meet the requirements prescribed in the job description and person specification hardly needs to be stressed.'(Tyson, 2007).Candidate is assessed by various selection methods and these selection methods are the decision making factors as well. 'Selection methods provide a systematic means by which information can be gathered about candidate and help predict their performance in that particular job' (Dale,1995).
'There are wide variations in recruitment and selection practices, reflecting an organizations strategy and its philosophy towards the management of people'(Bratton et al , 2000).Many organizations would do the traditional one-to-one interviews where as some may have day long assessment days where candidates are screened using different techniques. Farr et al suggest that any organization must consider the four important points before a selection program can be called sustainable and that is 1. Support of organization purpose, 2. Alignment with HR strategy. 3. Acceptability of governance and 4.effectiveness of process management (Farr et al, 2010). Support of organization implies that the selection activity must get a person that is the right fit for the organization. Daine Arthur(1995) describes 'right fit' is by identifying the past achievements and skills, willingness to use the skills and how? And lastly the intangible qualities. The integration observed by Delery and Shaw (2001) where HRM strategy links with the firms performance level clearly indicates that organization can maximize profit and production with the aliening its purpose with the HR strategy with and the aim of the business. (Askoh chanda HRM strategic intergration..)
The four elements form a part of the decision making factor for the choice of the methodology used for selection, but many authors point out two key factors that underlie any selection activity and that is reliability and validity.
The 'Reliability can be stated as consistency in the measurement or extent to which an instrument gives the similar results when subjected to same working conditions' (www.oppaers.com) 'According to CIPD reliability of a selection technique refers to 'Temporal or're-test' stability where the effectiveness of a selection tool is assessed by consistency of results obtained over time and Consistency - that is, does the test measure what it sets out to?'(www.CIPD.co.uk).Considering the factor that selection deals with humans and each person is different from the other it would be very difficult to obtain the same results and this is where the reliability is used to ensure the best selection tool is used in order to get good result.
Several different sorts of reliability test are used in selection technique research. Searle also points out the reliability is an issue that is increasingly becoming an matter of legal requirement.(Searle, 2003)
The first test of reliability that is Test-Retest Reliability 'can be done by getting the same person to complete the same measure on two different occasions.'(Searle, 2003).A central measure to this type of reliability is that the trait measured remains the same and does not change with time. The scores on the first test must be related to the scores of the second test in order for the measure to be reliable. If the scores are different then the measure does not stand valid. Reliability in this case is usually expressed as a correlation. 'The scores may be interview ratings or ability test scores or personality questionnaire profiles. Retest reliability is also used for work performance measures'(Cook, 1996).the second type of reliability used in selection is Inter-rater reliability. This is not a very popular and widely used measure for reliability but it still allows generating results. This method involves one candidate being assessed by two assessors and each gives scores and if the scores of the assessors do not match then at least one of them must be wrong. The author suggests that one of the reason of the method not being popularly used would be that one could never easily get two assessors to score the same hence the results that shall be obtained will more likely be wrong and will be disputed. The last measure that is used is Internal Consistency Reliablity. 'This is concerned with reliability through the internal consistency of a test, rather than the temporal change and learning of the test taker' (Searle, 2003).'Internal consistency reliability checks whether all questions are measuring the same thing'(Cook, 1996).Consistency can again be measured in two ways split-half testing and Kuder-ricahrdson reliability or also known as coefficient alpha. In split-half testing the results or the test itself can be split into two and the results from both can be compared. 'But, split half can create problem in that the test which is completed is only half to its correct length' (Searle, 2003).the other method which is the Coefficient alpha is much simpler to carry out. 'It is based on result of only one administration of the test. And this form of reliability is based on assumption that reliable test contain more variance and therefore more discriminating' (Searle, 2003). The reliability of a selection method as discussed above can be obtained by using various tests. All the measures give results in mathematical form and hence are easier to understand and predict. But, reliability measures are supported by the measure of validity.
Validity of a test is concerned with what is being measured. Kaplan and Saccuzzo (2001) define validity as 'the agreement between a test score and measure and, the quality it is believed to measure.' In other words, it measures the gap between what a test actually measures and what it is intended to measure. The gap can be caused by two particular circumstances, '1.desing of the test not suited to the real purpose intended and 2. The test being used for a purpose which it was not intended for'.(www.changing minds.org). Vernon(1963) , however added an interesting modification to this, suggesting' that validity may vary depending on the purpose it is put to'(Searle, 2003).There are four ways in which validity can be identified: face validity; content validity; construct validity and criterion related validity.
Face validity as the term suggest means that just by looking at the test the validity can be established. ' Face validity is especially important to test takers as they have invested time in trying to get the job and therefore want to feel that they have been assessed by something that is relevant for the role'(Searle, 2003). Searle indentifies that a possible problem that lies with designing test that pass the face validity that at times test-takers may get cues just by looking at the test, which may be irrelevant and distort their responses accordingly.
Content Validity 'pertains to selection procedures that test a sample of significant parts of a particular job that is- a demonstration that the content of a selection procedure is representative of important aspects of job performance' (Arthurs, 1998). Construct Validity is concerned with the construction of the test. 'Construct validity assumes that we can define and measure anything. The adequacy of the original definition and the availability of appropriate comparison measure is critical' (Searle, 2003). Validity helps in understanding of which personal attributes and personality traits are important in predicting the traits of the test-taker.
The reliability and validity as mentioned above form a very important part of the decision making process that is the choice of selection method. CIPD argues that 'validity, along with fairness, should be the overriding indicator of a selection method for high performance organizations and that it is important to obtain sophisticated data on validity in all its forms'. But, apart from this there are other factors as well that affect the choice of selection method and that is resource and cost constraints and the applicability of the technique. 'Selection techniques differ widely in their requirements and cost to support. Cost must be looked into from the perspective of development, production and usage' ( Roberts,1997).
3.2 Candidate Selection
According to CIPD, selection process has two sub-processes attached to it short listing, and assessing applicants to decide who should be made a job offer (www.CIPD.co.uk). Short-listing is reviewing the applications received to a quality and number that can be easily managed for the next stage that is the assessment stage. But, even before the author explores as to what takes place in a short-listing process it is vital to know what is the stage prior to it. The stage prior to this is that the candidate applying for a job.
Applications can be received in two ways, either through application forms or through a CV According to a survey done by CIPD, 'application forms were used in some shape or form by 80 per cent of the recruiters especially in larger organizations and the other alternative available is to allow candidate build their own CV. Many employers made use of both the application form and CV' (Taylor, 2005). The aim of both CV and a customized application form is to obtain as much information from the applicant in order to make the right decision. According to BIS ' CV provides a chance for the candidate to promote and sell themselves in a creative and unrestricted way, and an application form can provide a more accurate basis for comparison of candidates and help by asking candidates to provide appropriate and relevant information .'(www.bis.gov.uk)
Torrignton and Hall(1995) point out that 'application form initially were intended to form the basis of an individual's personal record but as the times and labour market has evolved it now forms part of a decision making activity'.Diane Arthur(1998) says that 'every organization must have their own application form whether paper or electronic, that reflects its own environment'. 'The design of an application form will clearly depend on particular situations and needs.'(Tyson, 2007). Application forms consist of set of questions which an employer will need to know about the candidate before inviting them for an interview. Alan Price comments that 'comparison of applications becomes easier if the information supplied is in a standard form (Price, 2007). For this particular reason employers develop application forms that help in achieving uniformity. Even though the application forms help in yielding al lot of relevant information from the applicant they suffer from certain drawbacks which are 'they could be time consuming, language barrier, too much information that might not be relevant (www.BIS.gov.uk).'
Another way of applying for jobs is through Curriculum Vitae or widely known as CV. CV's provide a creative unrestricted way of promoting oneself for a job position. Mark Freebairn(2007) in an article comments that 'CV is going to market you as best as possible'.() CV's do not have a set format but a general standard is observed and varies from country to country. 'A single CV can be used for many job applications' (www.BIS.gov.uk). A CV allows the 'applicants are able to tailor their application to their own strengths and are not restricted to fitting relevant information into boxes of pre-determined size' (Taylor, 2008). Cv's are widely criticized for information overload by applicants and at times information that may not stand true.
All applications and CV's received go through the short listing process or what is called as 'reduction' process .Reduction has the objective of reducing the pool of applicants to a manageable number by eliminating and rejecting unsuitable candidates' (Pilbeam et al, 2006). This is done by using job descriptions and person specifications which are drawn up during the recruitment process. This is the stage where the person involved in performing the reduction carefully bares in mind the law and equal opportunities and makes correct decisions.
Reduction is widely also known as screening of application. The applications can be screened in two ways either the traditional way or the modern way with use of technology.' Some of the traditional devices used to screen resumes no longer bear the weight they once did. These include the quality of the stationery, the design of the actual document, and the envelope in which the documents arrived' (Heathfeild, 2012).The traditional way which many recruiters still make use of is by developing competency charts or screening charts. As described by BIS a screening chart is' A screening chart enables you to mark which candidate meets which criteria. It allows you to go through each CV putting a score by each key criterion. You will be able to quickly see which candidates best meet your needs.'(www.bis.gov.uk). Hence once a job application is received the companies rate the applicants on the screening chart and the ones having higher scores are the ones that are then invited on to the next stage. The biggest drawback of the manual method is it being slow and is subject to human errors.
The use of online recruitment methods has given rise to online screening methods. 'Online recruitment can mean employers receive large numbers of applications from unsuitable candidates, so it can be helpful also to use technology to help manage the application forms'(www.cipd.co.uk). Hence once the application is received via internet the applications are run through the screening system and the screening systems then chooses the right applicants. Modern online screening tools include personality questionnaire, ability test, and self-test questionnaire. These all aim to establish the basic person skills requirement of any job. These are significantly used in areas where employers are trying to attract application from people who are less experienced says Stephen Taylor. One of the most popular known electronic software of screening application and CV is the 'Resumix'. 'Resumix does more than just scan and file applications, it is also a job analysis system'(Taylor, 2005).According to Mark Cook (1996) ' automated sifting systems can eliminate bias directly based on ethnicity, age or gender.' Other authors like Joyce Kennedy (2011) argue the 'these systems do nothing but search for key words'. This again leaves a question that how carefully the company setting up this system has selected what key words that could best describe the person who will be suited for the job role. It does pose a question of whether or not the talent has been lost as the candidate was unable to write a CV that will successfully pass the system test. 'The whole concept is highly questionable professionally as it introduces a strong element of chance into the short-listing procedure where good candidate is rejected simply because they have used one word rather than the other , while poor ones slip through'(Taylor,2008). Lisa Rowan, program director of HR, Learning and Talent Strategies,IDC explains that modern application screening systems or what are now called as ATS(application tracking system) not only look for key phrases but "It goes much further, (with the technology looking) at descriptive materials, and (the ATS) parsing things out like a human would when reading it."(Vass, 2009)
It would be difficult to say which system is best suited for any particular job position. Patrick Hauenstein , Vice President Navient Corporation(2005) points that any effective screening tool must be 'efficient, valid, legal defensibility, length, fairness and coverage'. He also goes on to compare manual Cv screening to automated CV screening 'Manual resume reviews do not perform well considering efficiency, validity, or fairness criteria and automated CV screening address efficiency concerns and reduce subjectivity that might adversely impact fairness, but still do not perform well when considering the validity criterion'.(Hauenstein, 2005)
Thus, recommending one system of application screening to be the ideal one for any selection process would not be suitable. Every company has to make a choice on the method to use as it also depends a lot on the position advertised and the resources a company has available for the screening process. All efforts must be made from the side of the applicant to overcome the first barrier of the selection process and move to the second stage of selection which is the candidate assessment.
3.2.2 Selection Techniques
As discussed earlier the very first part of an actual selection process is screening of the applications received. But, for many small companies in olden days the application forms were a basis of selection and the interview or one-to-one meeting would just be to formally. Philip Plumbley explains' well-constructed application forms are one of the best selection tool, as they compare like with like and the selector is unbiased with the presence of the person' (Plumbley,1991). But, one cannot solely rely on application forms as they do suffer from various drawbacks and one of the biggest being of language. Hence, companies moved on to `use what is also referred as the 'Selection by the classic trio'. This is the trio of application form, reference check and interview (www.media.wiley).
3.2.2.a Application forms
Have been discussed in the earlier section in quite depth and hence the author will proceed to the next selection tool which forms the part of the trio and that is interviews. 'The history of interviewing could be traced long back to the 13th century where the interviews were more like in form of interrogation, literature on interviews started appearing from the 1930's onwards and today there is endless data available on them'(Edenborough,2005).
'An interview is a specialized form of conversion conducted for a specific task-related purpose' '(Whetton & Cameron, 2002). The purpose of an interview is to fill the vacant position with the most suitable candidate. 'Interviewing is the most widely used selection tool but also the most abused and least reliable if left in the hands of the untrained,(Plumbley, 1991). Interviews are broadly classed as unstructured and structured interviews. This classification is broadly based around as the name suggest around the structure of the interview. Both the structured and un-structured interviews have sub-categories. To understand the sub-categories it is important to have a clear picture of what the two main categories are. Un-structured interviews are one in which 'Interview is without any set format but in which the interviewer may have some key questions formulated in advance. Unstructured interviews allow questions based on the interviewee's responses and proceeds like a friendly, non-threatening conversation. However, because each interviewee is asked a different series of questions, this style lack the reliability and precision of a structured interview. Also called non-directive interview' (www.business directory). 'The interview is unstructured in the sense that the candidate is encouraged to lead the interview and talk freely in response to the set of questions or a question' (Roberts, 1997).
The questions asked during an unstructured interview are not based on a job analysis and unlike the structured interview; follow-up and probe questions are encouraged (Blackman, 2002).
In most cases the unstructured interviews begin with the question of 'tell me about yourself' and the interviewer has to ask questions from the answer given by the interviewee. 'An unstructured interview is more open to interviewer effects than it's structured and semi structured counterparts. To become a skilful interviewer it takes knowledge and experience' (Minichiello et al, 1990).A question then arises about the reliability and validity of using unstructured interviews as a method of selection. Many researchers have conducted meta-analysis to predict the validity of the un-structured style of interviewing. One such research was conducted by Steven Crownshaw from the Department of Management Concordia University. Montreal (1988) found that 'structured interviews produced mean validity coefficients twice as high as unstructured interviews. The research also suggested that higher validity coefficients are associated with more reliable interviews and the use of formal job-analytic information in developing interview questions' (Cronshaw et al,1988) . Michael Companion and David Palmer from University of Huston also conducted a study aiming to review the structure in selection interviews also comment that 'previous review has concluded that structuring the se-lection interview improves its psychometric properties, they suggested that interviews can be improved by involving a structure and should be a future for research and practise' (Companion et al, 1997).
Structured interviews do not mean following a seven-point plan but it simply implies that questions are structured, judgments are structured, and certain phases of traditional interviews are sometimes eliminated(Cook,1996). In recent years several developments have occurred in structured interview techniques that are aimed to improve test results. 'There are several ways to structure an interview but all would primarily involve systematic and pre-determined rules of observation and evaluation' (Stephan et al ,1995). Structured interviews, regardless of content, are more valid than unstructured interviews for predicting job performance criteria .When the content of the interview is job related, structured interviews are still more valid than unstructured interviews (McDaniel et al , 1994).
'A major ¬nd in interview research a few years ago was that interviewer judgments based on structured interviews are more predictive of job performance than those from unstructured interviews' (Macan, 2009). In fact, 'many quantitative and qualitative reviews of the employment interview research have concluded that adding structure to the interview process can enhance the reliability and validity of interviewer evaluations'. (Conway et al., 1995; Huffcutt & Arthur, 1994; Huffcutt & Woehr, 1999; Posthuma et al.,2002). Schmidt and Zimmerman (2004) explored a 'measurement-error explanation'. They hypothesized that structured interviews show higher validity for predicting job performance compared to unstructured interviews because structured interview ratings are more reliable. (Macan, 2009)
'Despite the evidence showing that interviews containing high levels of structure can be valid predictors, surveys show that managers, HR professionals, and organizations infrequently use them' (Klehe, 2004; Lievens & De Paepe, 2004; Simola, Taggar, &Smith, 2007; Van der Zee, Bakker, & Bakker, 2002). 'Conceptualizing interview structure as a continuous variable with various levels along two dimensions, most HR professionals have been reported using interviews with a moderate degree of structure' (Lievens & De Paepe 2004). 'Interviewers typically engaged in interviews in which they had identi¬ed the topics beforehand (i.e., moderate level of question standardization) and rated candidates on multiple established criteria (i.e., moderate level of response scoring). One has to wonder to what extent research studies conducted'(Macan, 2009).
The sub-categories of interviews are, one-to-one interview, exploratory interviews, telephone screening interview, video screening, departmental interview, panel interviews, and peer interviews. The above interview techniques are based on the requirement of individual companies, the resources available to conduct a particular interview and also the requirement of the job.
A poorly conducted interview may leave the candidate with an unfavourable impression of the organisation that they are likely to share with other potential applicants and customers. It is good practice to give feedback to candidates following an interview. (www.cipd.co.uk) Anderson and Shackleton (1993) draw on a wide variety of studies to summarise the reasons why interviews have been criticised in this way. These include: self-fulfilling prophecy effect, stereotyping, halo and horns effect, contrast effect, similar-to-me effect and personal liking effect.
3.2.2.c Assessment Centre
James Larue defined the assessment centre in 1989 that 'An Assessment Centre consists of a standardized evaluation of behaviour based on multiple inputs. Judgments about behaviour are made, in major part, from specifically developed assessment simulations. These judgments are pooled by the assessors in a meeting among the assessors or by a statistical integration process. In an integration discussion, comprehensive accounts of behaviour and often ratings of it are pooled. The discussions result in evaluations of the performance of the candidate on the dimensions or other variables which the assessment centre is designed to measure. Statistical combination methods should be validated in accord with professionally accepted standards'(Larue , 1989 ). ' The basic principle requires that candidates, usually for management positions in organizations, go through a series of individual and group tests and exercises in one concentrated period while being evaluated by a group of assessors'(Howard,1974).
Generally in assessment centre exercises and in the final evaluations each participant is evaluated by more than one assessor. Accordingly, 'inter - rater reliability becomes a matter of some importance, in addition to the reliability of individual measures' (Howard, 1974). 'Over the past three decades, assessment centres have been steadily gaining popularity with organizations worldwide, for the purposes of employee selection and development' (Eurich et al,2009; Joiner, 2002; Spychalski et al, 1999). 'Assessment centres appear to exhibit both content- and criterion-related validity while simultaneously lacking construct-related validity'(Cahoon,2012).
3.2.2.d Psychometric Testing
According to The British Psychological Society definition of a psychometric test: 'a psychological test is any procedure on the basis of which inferences are made concerning a person's capacity, propensity or liability to act, react, experience, or to structure or order thought or behaviour in particular ways'(www.agcas.org).
According to Keith Coaley psychometrics are designed to do measurement; in fact the term is abbreviation for psychological measurement' (Coaley, 2009). Psychometric test are used in number of fields, such as schools, colleges and at work. Here the authors shall focus on the use of psychometrics test by employers to select the best candidate. 'Psychometric test give those taking the test and opportunity to respond to a series of items or events that relate directly or indirectly to a particular area of behaviour' (Edenborough, 1999).
'Most tests are designed and developed by occupational psychologists and are accompanied by detailed manuals providing the data to establish the reliability of the test and the normative information against which test results may be compared' (www.cipd.co.uk). Each psychometric test aims to measure different things. In corporate worlds are used for Selection of candidates to jobs Personal development/identification of training needs/staff development Careers guidance Building and developing teams (www.agcas.org). Toplis et al (2005) say that 'A well designed psychological test will have six properties 'objective and standardised measurement, the test items are ordered by level of ability, objective scoring, interpretation using norms, technical manuals, and objective evidence of validity'.
The various types of psychometric test available are personality measure, ability aptitude and attainment, interest inventories, educational testing, and clinical testing. Personality measure aims to asses an 'individuals behaviour and the way it is organised and co-ordinated when he or she interacts with the environment '(topils et al, 2005). But, theorists have always found it challenging to describe personality. However in general they define as 'any significant behaviour that distinguishes one person from the others and can be consistent over situations and time'(Coaley, 2009). Personality test include, self-report personality questionnaires, interest questionnaires, emotional intelligence questionnaire, values questionnaire, work behaviour questionnaires. Personality tests help in gaining an insight into characteristic behaviour of one person in relation to another or to a group of others (Edenborough, 1999). Ability, aptitude and attainment are test of general intelligence. 'the term aptitude is usually reserved for those tests directed at predicting whether skill in a particular area can be acquired and hence these tend to be more job specific, whereas ability tends to be reserved for less job-specific measures , though often still job-related intellectual characteristics( Edenborough, 1999). Test of attainment help is establishing the degree of knowledge or skills a person has acquired at a particular point in time. Commonly used ability, aptitude and attainment test include verbal, numerical test. Interest inventories are used for career development purposes. This helps in establishing the career interest of employees and helping them to achieve it by providing training and development. Educational testing is commonly used 'to describe assessments of children and teenagers in connection with the educational process.'(Edenbororugh, 1999).Clinical testing is used on medial level to help determine anxiety, depressions or any small damage to brain that might have occurred.
Like all the other selection to tools it is important to ascertain the validity of psychometric tests. It is difficult to obtain the predictive validity of the test as the employees would have gained skills during their current employment which may affect their performance in the test. The test does not support content validity as it these tests are not comprehensive. Psychometric test do have face validity. Example being the 'inkblots' test if not well depicted can cause irritation in the test takers and can affect their attitude not only to the specific test but to the entire test procedure(Toplis et al, 2005). Thus, it must be ensured that any psychometric test used by companies does come with its validity and reliability to ensure suitable results.
Psychometric test suffer certain drawbacks such as 'Bias between test takers for example gender bias, poor application of tools, inadequate job analysis, wrong usage of tools, Words defined differently by developers causing confusion, misinterpretation of results by users. Not reading the test manual properly.(www.changing mind.org)
Founders Larry Page and Sergey Brin met at Stanford University in 1995. By 1996, they had built a search engine (initially called BackRub) that used links to determine the importance of a individual webpages.(Google.com). The Google story is at the heart of Google's success. The success of Google has been phenomenal. 'Today it indexes over 8,000,000,000 web pages, ready to be trawled for information when we're looking for it. It has versions of its website operating all over the world, and in many different languages. Google has a big team of people trying to outwit the people who are trying to outwit Google' (marshal Cavendish).
The selection process at Google is very different from that of its competitors like Apple and Microsoft. Much that is known about Google points out that mindset is especially important because at Google, being liked is as important as being talented. Google prequalifies candidates by skill and mindset before setting up an interview. (Shipman, 2006). 'Each candidate is interviewed not only by the line manager but also have at least two of their would-be peers and the people they would be managing' (Dee, 2010). The author finds this process may lack the inter-rater reliability as the group may score the candidate differently. All the above is what has been disclosed by various news paper articles and journals. Google has also disclosed their recruitment process on their company website. Google says '. Our process is pretty basic; the path to getting hired usually involves a first conversation with a recruiter, a phone interview and an onsite interview at one of our offices. But there are a few things we've baked in along the way that make getting hired at Google a little different'. Google also comments that each candidate is likely to be interviewed by at least 4 to 5 Googlers and the aim behind the interviews would be to establish leadership qualities, role related qualities ,how the candidate thinks and finally googleyness.(www.google.com)
In the following part of the project the author shall be analysing the interviews that were conducted with the Google employees and try to evaluate the process.
The following research will enable the evaluation of the current employee selection technique at Google Inc, Mountain View California. And, ultimately basis for any recommendations and areas that the company can do further research into.
The research process will consist of following:
Review of literature on various selection methods and process used currently around in business world.
Review the literature available about the company, from company website and trade journals.
Conducting semi-structured interviews of current employees of Google.
Analysis of the interview data and establishing any potential development or further research areas if any.
The research is aimed at the evaluation of the employee selection technique at Google Inc, and involved in getting in-depth information from a limited number of people. Hence, qualitative research was most appropriate. 'Qualitative research is a research that addresses business objectives through techniques that allow researcher to provide elaborate interpretations of market phenomena without depending on numerical measure'(Zikmud , 2003). A thorough review of the literature will provide a good base for evaluation of the selection process at Google.
In words of Patton(1980) 'the purpose of interviewing is to find out what is in and on someone else's mind. We interview people to find out from them those things that we cannot directly observe'. In this research also the researcher aims to establish a dialogue with the employees at the company. For, this purpose also structured interview would not have been the ideal choice as the structured interviews limit the answer that the interviewee gives. Hence, the researcher conducted semi-structured interviews. 'Semi-structured and in depth (unstructured) interviews are often referred to as qualitative research interviews' (King, 2004).
Hence, the researcher developed a set of questions based around the objective of the research. These questions were reframed according the flow of the conversation. As Saunders's et al(2007) explain that 'in semi-structured the order of the questions may vary as well as there may be a need to ask additional questions depending on the flow of information'.
In qualitative research a number of considerations may influence the selection of a sample such as ' the ease in accessing the potential respondents, your judgement that the person has extensive knowledge about the situation, and how typical the case is of category of individuals or simply that it is totally different from others'(Kumar, 2011). In this research the sample choice has been affected by two factors, one being the ease in accessing the respondents and secondly the knowledge of the situation.
Hence the sample is drawn from two section of the population. Sample A consist of 10 managers who are on interview panel and sample B is drawn from the 10 employees who are not involved in recruitment process.
As discussed earlier, the questions will be based around the objective of the research and shall be formulated in a semi-structured format with open ended questions. Open ended questions are flexible and allow the researcher to probe to get deeper understanding and avoid any confusion. 'The major problem with semi-structured interview is coding or content analysing the response' (Burns, 2000).
Informal observations help in better the understanding one has about the organization and its processes. The researcher had a few opportunities to make informal observation while the visit at the Google Inc London office and The California office. The observations did not directly help me with my answers but they did help me to assimilate the culture of the company, which indirectly helped me in conducting the interviews in a much better way and also interpret the responses from the interviewee.
Data recording methods
Data was recorded in written form. Notes were taken during the interview process as Sekaran comments that 'the interviewer should not rely on memory, because information recalled from memory is imprecise and often likely to be incorrect' (Sekaran,2003). The interview was not audio recorded in order to create any doubt in mind of the responder about the use of the recording. As Sekaran(2003) explains 'taped interviews might bias the respondents' answer because they know that their words are being recorded and their anonymity is not preserved in full'.
Hence the researcher rejected the idea of recording the interviews; this not only enabled to get most information from the respondent but also helped in establishing a good rapport. The interview notes were reviewed by the respondents and their consent was taken in order to ensure that they are happy with what notes have been taken and
Reflective statements as the word suggest is reflection on the work done. During this research the researcher faced many challenges and during the writing of the starting from the planning stage, until the end.
During the planning stage of the project it was projected that the interviews would be conducted by the end of July along with their analysis. But, this seemed rather difficult to get the time arranged with the employees at Google and getting the data from them. Time was also spent in collecting the data from previous literature as the process took far too much time as the researcher expected it to take.
The second challenge that the researcher faced was extracting the information from the respondents. The respondents were very cautious in providing the information and answering the questions as the questions were all related to a very important process of the company that distinguishes them from their competitors. Hence they were making sure that they do not give out any information that could be counted against them.
Time constraint can be avoided by devoting more time in the planning stage and making sure that there is a contingency plane. 'No one ever sticks to their plan perfectly, and you can't predict all the things that might intervene, so build in some extra time for '(www.reading.ac.uk.). The author lacked a contingency plan and was hoping that all the respondents would adhere to the decided timeline, and failing that the interview stage extended up till august first week. Another learning here was that make sure that one keeps the interviewees in a constant communication , rather than dropping them emails a week prior to interview date. A constant touch with the respondents enables to build rapport and manage time as the respondents when kept in touch are aware of the progress of the project and are mentally preparing themselves for the interview stage accordingly.
The second challenge that the researcher faced was that the respondents were over cautious in providing information. One of the learning which is applicable here as well is always keeping the respondents in constant touch. This helps in building rapport and trust amongst each other and the respondents feel more comfortable in talking. As Martin Higham explains 'establishing rapport really means getting into communication '(Higham, 1997).
These are the two prominent learning's that the researcher has got from this research project and will aim to improve on them in the near future.
Analysis of Data
The analysis of the data received from the employees of Google Inc. has been consisted of managers and senior level employees who are involved in the decision making process of whom to select and also few of them are ones who have actually frame the questions for the interview.
The aim behind every selection interview is to place the right candidate in the right position. As described by Margaret Foot and Caroline Hook 'is to obtain employees who will be productive and committed members of staff, working and developing to their full potential' (Foot et al, 2005).
The selection interview at Google as the explained by Google is that it does not aim at just filling the position but filling it with someone who would stay with the company for a long time and is not just good at their job but can do lot of things. Most companies do have a similar aim of hiring someone who would stay with the company and will match the future needs of the company. 'long term focus should be apparent .the organization is not simply looking to fill the space , but is thinking long term about recruiting people with attributes that will be required increasingly in future'(Taylor, 2010).
The aim of hiring for long term really reflects in the employee turnover and the overall company turnover Google has. The interviewees claimed the turnover to be between 1.5 to 2 percent at the Google California head office and the overall company turnover was expected to rise by 1% in first quarter of 2012 (www.investor.google.com). This staff turnover is a fairly low number considering the size of the company and the overall economic climate of recession. There could be many potential reasons behind this, but since the aim of the research is centred on the selection process the author shall focus on it.
Hence , the question here to ask is that is this turnover low just because the hiring process is designed in such a way that it hires only the employees who stay with company longer, or are the perks that the employees get are good enough to retain them in the company for long. When asked the reason for the low turnover the common consensus was that the perks that are associated with a job a Google are very lucrative. Abhinav explains : ' New hires begin with three weeks' vacation during the first year, employees get free breakfast, lunch and dinner at premises, once a month employee can dine in any of the Google's restaurants with family for free and finally there are no sick days at Google , if you are not feeling well you just stay home'.
The perks offered by Google cannot be matched with any other company. This points clearly towards the effective employee retention and talent management the company has and no possible link can be established between the low staff turnover rate and the selection technique.
But, what really makes a Google selection process. The Google selection process from the outside seems as ordinary and simple as any other company. The candidate applies to Google and the CV is run through the ATS (application tracking system) that does the initial screening and there after the screening the information is collected about their school and college and is matched with an existing employee at Google who went to same school and the employee is asked to give an internal reference, following which if it is positive the candidate, then attends a telephone interview. The aim of trying to match the external profile with an internal candidate points towards what Margaret Dale describes as 'stereotyping a way of saving mental energies and the trying to use this can be grossly misleading , leading to false conclusion'(Dale, 2003). On the other hand Gareth Robert says' there are negative and positive stereotyping. The positive stereotyping is where people of certain background or experience in common with others employed, but of no casual relationship to performance, are deemed suitable'(Roberts, 1997) The author still finds the preliminary process of selection at Google quite questionable as matching the profile is not as bad as asking the internal employee to provide a reference which may be subject to personal bias.
The next stage being the telephone interviews which Michele says 'the number of telephonic interviews one may go through could range anything from being one to 5 in the very same day'. Telephonic screening is being used by many recruiters since a very long time and only thing to bear in mind which Daine Arthur points out is to keep ' style and substance separate'(Arthur, 1998). It means unless the job role demands the person to have excellent communication skills, any lack in the telephone screen in terms of language should not be classed as a negative towards the candidate.
Following the telephonic screening the candidate goes through a GLAT test. The Google Lab aptitude test, these test as Taneja explains' are created by few or our highly skilled staff members and aim in establishing your mathematical, computers and Google culture'.
Every candidate's walks in for an on-site interview with a GLAT score (a sample GLAT test is attached in Appendix 1). The GLAT score may not be a true representation of the actual person, example a person applying for the job of a reception position may not be good at maths and hence may not score well on the test and hence there arises a question about the validity of the test. But, Google justifies that the test are valid as they are necessary for the job to be done.The researcher was able to summarise from the conversation that GLAT has criterion-related validity , content validity and construct validity. This becomes more clear by moving further ahead in the evaluation of the entire process as GLAT does not stand alone as a determining factor, it is supported by the face-to-face interview to make the final selection.
The last stage, onsite interview comprises of questions that are leadership oriented, analytical questions, job role related questions and technical if one is applying for a technical or IT role.
The entire process described so far seemed to be similar to that of any other company then how is the Google process different from that of its competitors? A general answer was that our interview questions make us different; we focus on finding real Googlers. But, Jose went on to explain this in little bit more depth. He said 'Google focuses a lot mathematical and analytical ability, and hence all our questions are framed accordingly. There is no noticeable difference between us and our competitors except our GLAT which is especially designed for Google'. Thus, this statement helps in establishing the validity further more as the test revolves around the need of Google. The response received from the respondents has not provided enough data to proceed with further in testing the validity of the test.
The next question in line is how the hiring decision is made. A fact that is known about Google is that Larry and Page sign off every single Google hire. But, apart from this Google does what they like to call as 'consensus hiring'. Claudia goes on to explain what consensus hiring is ' we let the peers participate in the selection process, so in any interview about six people performing the same job role would be involved in the hiring process'. The entire group of responders agreed to the reasoning behind this process which was that 'employees feel they are involved in the process and their voice matters'. The researcher believes that is a good way of improving the psychological contract of the employees as this gives them the ownership of the company and also this indicates towards a good employee relation practice is in place and Google would be a good example of sophisticated modern style of management where employers consult their employees (.....)
Google is known to be taking far too long. In a computer journal article Eric Lai (2006) wrote 'Google's arduous and secretive hiring process has started outweighing the glamour and perks it offers' (Lai, 2006). He also commented that the company is trying to streamline the process. The researcher came across several other internet articles which commented on the lengthy selection process at Google. Hence, the researcher decided to get the real picture from the respondents. The response that the researcher received across the board was that ' the process does take a lot of time, any given candidate would go through at least 5 interviews before they are placed and these could be spread over a month'. The respondents themselves led towards answering another question which was 'if there has been any change to the recruitment process in last on year'. Melanie explained' the process has changed in the last on year, we have streamlined it to four to five interviews, whereas in past people could face up to 14 interviews which could take more than a month'.
Moving to the next question which builds a myth in the tech world about Google's selection process and that is the use of advanced level math questions. The researcher here is aimed not only to find out what the reality was and what the reasoning behind it is. The respond received across the board was that yes Google does focus on judging a candidates mathematical ability and their presence of mind. The reason behind it is that 'no matter what position you apply for you would need numbers in your day to day life but if you are talking about are GLAT which is taken only by candidates apply for technical position'. The researcher in the beginning of the analysis has tried to determine the validity of GLAT and with the limited amount of information that could be gathered from the respondents came to the conclusion that GLAT does stand as a valid test. But, there could also exist a stereotyping effect here from the owners of Google Larry Page and Sergey Brin. Both them coming from mathematical background and hence when they started the company they wanted people like them to be a part of it.
The last question was aimed at depicting the effectiveness of the entire selection process .i.e. employee performance. The response to this question is that there are no specific performance measure tools in place at Google. 'Each employee picks up a project that they are interested in doing and they work towards it. The relation between and employee and manger is of a 360 degree feedback and the employee is as welcome to speak their side as much as a manger is. Also if an employee at the end of a year has not finished what the project they took they are not marked down for it, on contrary the manger is keen to see what developments have taken place since they took over the project'. Thus, it was very difficult to say for the respondents as to whether or not the people they selected performed as they hope they would.
' Google's ability to quickly develop new services and capture market share has made them leaders in the marketplace and forced less nimble competition to play catch-up'(Tharp,2009). To keep the position, Google needs to hire the right people who can contribute to the success of Google as a company.
The current selection process at Google may seem very simple when observed from the outside as a process, but it is in practise a very complicated and advanced process. It is a good mix of all the selection tools that are currently available whether it is telephone screening to advance level of psychometric test.
There is definitely a hype created about the Google recruitment process that is being very challenging and hence it still attracts many people to constantly apply for job. Though the process has its own positives and very few negatives, the researcher has in the next section has attempted to highlight out some potential research and development areas.
The researcher aimed in the project to evaluate the selection process at Google. The following recommendations are made based on the literature reviewed and the findings from the unstructured interview.
The first recommendation is in relation to the profile matching that Google does. There is a lot of ambiguity in regards with the validity of the entire process. As mentioned earlier as well that there are arguments in favour of using this technique and arguments against it as well hence the company must adopt another technique of preliminary screening that is reliable enough and is not subject to personal bias, which the current technique is highly likely to be. Google has reasoning behind this process which Karp explains 'this allows recruiters to tap employees who best understand the demands of the jobs and the nature of the culture in assessing the fit of potential hires. It allows current employees to build the community--even if they are not part of the formal interview process' (Wright, 2009). But, the author still suggest that the process is does suffer from the drawback of personal bias and its use must be re-evaluated as the candidates who do are unable to secure good internal reference may challenge the process of being dominated by the view of the internal employee about them. The best suited system would be if Google is keen to obtain references from in the preliminary stage the most suitable would be to perform an external reference check.
The second recommendation stems out from the time spent in the entire selection process. The process as explained by the respondents earlier was a 14 round interview which has been reduced to 5 rounds but even then it