A Study of E-recruitment Technology Adoption in Malaysia

A study of e-recruitmenttechnology adoption in Malaysia

David Yoon Kin TongFaculty of Business & Law, Multimedia University, Melaka, Malaysia

Abstract

Purpose – The purpose of this paper is to examine the employed jobseekers’ perceptions andbehaviours of third-party e-recruitment technology adoption in Malaysia.

Design/methodology/approach – Using the validated modified Technology Acceptance Model(TAM) without the attitude construct as the core research framework and identifying PerceivedPrivacy Risk (PPR), Performance Expectancy (PE), Application-Specific Self-Efficacy (ASSE), andPerceived Stress (PS) as key external variables that form the research model for the study ofe-recruitment technology adoption.

Findings – The results identify few key determinants to this technology adoption. Moreover, theweak evidence of the behavioural intention indicates that e-recruitment has not replaced some of theconventional recruitment methods.

Practical implications – The study implies that the third party e-recruiters’ policy makers andhuman resources practitioners need to improve the e-recruitment system and services to attract these“passive” talented groups of candidates for employment.

Originality/value – The paper provides an insight for human resources practitioners on theeffective use of third-party e-recruitment service provider and the strategy to attract employedjobseekers for employment.

Keywords Modelling, Online operations, Recruitment, Jobs, Malaysia, Sampling theory

Paper type Research paper

IntroductionIn human resource management context, recruitment is a process of sourcing andacquiring the right applicants to an organization. Essentially, the process involvesseeking and attracting a pool of qualified applicants using various feasible recruitmentmethods. The conventional recruitment methods used by organizations consist ofcontacting friends or employee referrals, engaging executive search, using newspapersclassified ads, and others. Whenever there are changes in company’s policy,technology, location, mergers, acquisitions, de-mergers, and employees’ resignation,this process continues to take place periodically to add, maintain, or re-adjust theirworkforce in accordance to the corporate and human resource planning (Tyson andYork, 2000; Cascio, 1998).

As global competition persists and industries becoming more skill intensive, therecruitment of talent workers becomes essential (Tong and Sivanand, 2005), andattracting the right applicants at the right time is getting tougher than ever. The use ofconventional recruitment methods no longer suffices and timely to attract sufficientpool of qualified applicants. Many organizations have turned to adopting sophisticatedrecruitment strategies or combining various recruitment methods to attract them

The current issue and full text archive of this journal is available at

www.emeraldinsight.com/0263-5577.htm

The author thanks the anonymous reviewers for the comment on earlier draft.

The study ofe-recruitment

technology

281

Received 14 April 2008Revised 30 June 2008

Accepted 25 July 2008

Industrial Management & DataSystems

Vol. 109 No. 2, 2009pp. 281-300

q Emerald Group Publishing Limited0263-5577

DOI 10.1108/02635570910930145

(Tong and Sivanand, 2005). For example, by combining newspaper ads with executivesearch, or employment agencies, and others for recruitment; but this only adds to theincreased of recruitment costs per hire (Pollitt, 2005, 2004).

In early-1990s, with the advancement of internet technology, many have witnessedthe transformation of the conventional recruitment methods to online recruitment(Joyce, 2002). Some corporate companies even use their web sites to recruit people whileothers capitalized this change to become e-recruitment service providers (Dixon, 2000).This latter “third-party” e-recruitment business thrived to become the second mostpopular online business besides booking airline tickets, in United States and Europe.This business trends was later caught up in South-east Asia and Asia Pacific (Labanyi,2002; Galanaki, 2002; Fisher, 2001; Gomolski, 2000).

The third-party e-recruiters provide services to companies who are interested to usetheir web sites for job advertisements and viewing potential applicants’ postedresumes at a fee lower than most conventional recruitment methods. Most e-recruitersprovide free services to applicants or jobseekers to post their resumes online in theirdatabases (Galanaki, 2002). With this free posting, the growth of resumes is inevitable.Millions of resumes are posted to famous e-recruitment web sites, becoming a truemarket; uncontrolled and unconstrained by geography (Cappelli, 2001).

With the rapid growth of the third-party e-recruitment web sites, this has altered theway jobseekers are looking for jobs, and the way companies are recruiting them butlittle is known of their effects on the labour market. Accordingly, in this study, it leadsto the reviewing and understanding of the job search and information technology (IT)literatures. Combining these literatures, this study uses the highly validatedTechnology Acceptance Model (TAM) developed by Davis (1986) as the researchframework to analyse and understand this considerably new technology adoption forjob search. The spread of this e-recruitment business to South-east Asia, particularly inMalaysia seemed desirable to test the local jobseekers’ perceptions and experiences one-recruitment utilization.

This paper is organized and presented as follows. First, the literatures reviewdiscusses the job search trends and the conceptual model and hypotheses. Second, theresearch method is provided. Third, the results are presented. Next, the discussion andimplications, and limitations of the research findings are discussed. Finally, theconclusion is drawn and recommendations for further study suggested.

Literature reviewIn the job search literature, Quint and Kopelman (1995) using the Job Search BehaviourIndex (JSBI) asked respondents (59 percent’ employed, 37 percent currently seeking foremployment) to indicate ten dichotomous items (yes or no), whether they had engagedin ten different job search activities over the past year, predicted job acquisitionsuccess is positively related to the level of job search behaviour. That is, for jobacquisition success, jobseekers’ application must demonstrate three variables:

(1) effortful job acquisition behaviour (motivation);

(2) possession of the requisite job-related knowledge and skills (ability); and

(3) an appropriately focused job acquisition strategy (role direction).

Comparing Quint and Kopelmans’ research, Mau and Kopischke (2001) also used tendifferent job search methods on college graduates (N ¼ 11,152) to find out the

IMDS109,2

282

percentage of students using any of the method(s) regarding their job seekingbehaviours and outcomes. In their research, race and sex differences among the jobsearch strategies were used. The variables include the number of job interviews and joboffers, annual salary, and job satisfaction were examined. The results indicatedsignificant race and sex differences in job search methods used. However, there were nosignificant differences in number of job interviews or job offers regardless of race or sex.

Comparing Quint and Kopelmans’ research, Meisenheimer II and Ilg (2000) alsoused ten items job search methods, which they surveyed on employed wage and salaryworkers. However, they termed these methods as active methods used by respondentsfor new job search. Their study first examines whether any trends have emerged in theproportion of workers actively seeking new jobs (job search rate) in February 1995,1997, and 1999. Then, the focus turns to how different characteristics of workers –such as sex, age, earnings, health and retirement benefits coverage, educational level,tenure with current employer, job security, occupation, industry, and unionmembership – relate to workers’ likelihood to seek new jobs. The analysis does notinclude employed jobseekers who are self-employed workers, wage and salary workerswho were looking for a second or additional job, and jobseekers that used only passivemethods to search for a new job. Passive job-search methods include merely readingthe want ads or attending a job-training program or course.

The findings indicate that age appears to have a stronger relationship with thelikelihood of searching for a new job. Workers ages 16 to 24 were more actively seekingfor employment opportunity than persons’ ages of 25 and older. However, as the labourmarket tightened over the 1995-99 period, the likelihood of young workers to seek newjobs declined, particularly when compared with the job-search rate of workers age 25and older.

Comparing Quint and Kopelmans’ research, Kuhn and Skuterud (2000) examinedthe frequency and incidence of internet job search among US workers, by race, gender,and other demographic characteristics, the location of the job search (from home, fromwork, or from other access points), and the relation between internet search andtraditional job search methods. The internet job search data are from a specialsupplement to the December 1998 Current Population Survey (CPS), which askedrespondents about computer and internet use. The traditional job search methods arealso from the monthly CPS used by the Bureau of Labour Statistics (BLS) to determineif a respondent is an active jobseeker. However, the CPS uses nine instead of tentraditional methods for the study.

In Kuhn and Skuterud (2000) study they cautioned that there is a possibility ofoverlap between search for a job via the internet and the traditional methods outlinedin the CPS. For example, unemployed jobseekers who say they “contacted employersdirectly” may have done so through the internet, perhaps submitting a resume viae-mail (internet search) or they may have actually mailed or personally delivered acopy of the resume to potential employers (traditional search). Kuhn and Skuteruds’research also indicated that although unemployed is more common in usinge-recruitment for job search, about seven per cent of employed workers are usinginternet to search for new jobs. Table I illustrates the summary of the comparison ofjob search methods discussed.

Overall, comparison between the questions used by these researchers indicated thatthe traditional job search methods have not changed much since Kopelman et al.


technology

283

introduced the JSBI in 1992, except the Kuhn and Skuterud (2000) study. However, atthe time of Kuhn and Skuterud’ study, e-recruitment was still at its “infant”development stages, which Byars and Rue (2000) highlighted that it seems safe to saythat the research has not identified a single best source of recruitment for recruiters yet.In addition, the digital divide among jobseekers might be an issue to earlye-recruitment study. However, since then, the growth of online population is inevitableand its growth is closing the digital divide gap (Pastore, 2002). This impliese-recruitment will grow concurrently with the online population.

While e-recruitment seemingly paves the way to become future recruitment method,and is highly likely to become jobseekers platform for job search, it is logically toreview the specific jobseekers perceptions and behaviours on this technology usage.As Peter (2001) pointed out that generally, there are two categories of jobseekers: activeand passive. Passive jobseekers are those employed jobseekers that already have agood position, but will apply if they see another job of interest, whereas activecandidates may include the dissatisfied, less employable jobseekers and passivecandidates are of higher quality than active candidates. In this study, the authorsought to examine the employed jobseekers which represent the passive group.

Conceptual model and hypothesesIn IT literature, the TAM is the most influential model use to measure technologyacceptance. This model is the extension of Ajzen and Fishbein’s Theory of ReasonedAction (TRA), by Fred Davis and Richard Bagozzi (Bagozzi et al., 1992; Davis et al.,1989) to explain the computer-usage behaviour. The main purpose of TAM was:toprovide an explanation of the determinants of computer acceptance that is generally,

No Mau and Kopischke, 2001 Meisenheimer II and Ilg, 2000Kuhn and Skuterud,2000

1 Send resume Sending out resumes Send resume2 Campus job placement office Interview through a school or

university employment centreContact schoolemployment centre

3 Look through want ads Answering ads - Not stated -4 Ask friends/family/professors Asking friends or relatives about

available jobsContact friends orrelatives

5 Attend recruiting fair Contacting an employer directly Contact employerdirectly

6 Do volunteer work in field - Not stated - - Not stated -7 Unemployment office Registering at public employment

agencyContact publicemployment agency

8 Contact head hunters/employmentagency/professional recruiter

Registering at private employmentagency

Contact privateemployment agency

9 Place own want ads Placing ads Placed or answered ads10 Subscribe to trade journals Checking union or professional

registersCheckedunion/professionalregisters

11 - Not stated - Filling out applicants Fill applications12 - Not stated - - Not stated - Used other active search

methods

Sources: Mau and Kopischke (2001); Meisenheimer II and Ilg (2000) and Kuhn and Skuterud (2000)

Table I.Comparison of job searchmethods

IMDS109,2

284

capable of explaining user behaviour across a broad range of end-user computingtechnologies and user populations, while at the same time being both parsimoniousand theoretically justified (Davis et al., 1989, p. 985).

Davis’ (1989) original TAM model uses five constructs that consist of perceivedusefulness (PU) construct as “the degree to which a person believes that using aparticular system would enhance his or her job performance,” and perceived ease ofuse (PEOU) as the degree to which a person believes that using a particular systemwould be free of effort.

Many researchers’ empirical studies have replicated and tested the model underdifferent conditions for TAM’s extended variables as general measures by explicitlyincluding IT acceptance variables, such as extrinsic and intrinsic motivators (Igbariaet al., 1995; Davis et al., 1992), computer self-efficacy (CSE) (Agarwal et al., 2000; Lopezand Manson, 1997; Compeau and Higgins, 1995), social influence, and among others(Morris and Dillon, 1997; Malhotra and Galletta, 1999; Said and King, 1999; Mathieson,2001; Klopping and McKinney, 2004; Ma and Liu, 2004). However, Davis et al. (1989)TAM postulates that PEOU and PU are of primary relevance for computer acceptanceand Davis (1993) noted that PEOU might actually be a prime causal antecedent of PU.

In this study, the author addresses the gaps by replicating and testing modifiedTAM without the attitude construct and identifying key constructs as externalvariables that justifies the study. Drawing from the past validated studies of theexternal variables, the rationale forms the research questions, hypotheses, andresearch model for e-recruitment as technology for job search method, which paststudy was unanswered.

Perceived usefulness (PU)Effective e-recruitment service providers often support jobseekers with comprehensivejob information and some with career enhancement tools in the web sites, which theycan conveniently assess for their career plan (Tong and Sivanand, 2005). This isusually available at the click of the career zones that offer occupational information,which includes effective resume writing, continuing education, salary information, andinterviewing information, featured career articles, and self-assessment to guidejobseekers (Rosencrantz, 1999) in which the traditional newspapers ads do not offerthis. Jobseekers rely on job information available to them when applying for jobs(Fountain, 2005). Perceiving system usefulness as antecedent of e-recruitmentutilization, such as using these information and tools to enhance the effectiveness of jobapplication, would draw the attention of many employed jobseekers into adopting thetechnology for job search.

Perceived ease of use (PEOU)In general, if a system is easy to use, less effort is required by the users, therebyincreasing the likelihood of usage. Conversely, a complex system is difficult to use areless likely to be adopted since it requires significant effort and interest on the part ofthe user (Teo, 2001). Similarly, in the e-recruitment context, jobseekers would prefer thesystem if it is easy to use compared to other methods of job applications.

In the study of Web acceptance, Sanchez-Franco and Roldan (2005) found therelationship between PEOU and PU was significant and positively related. In thee-recruitment context, the easy to use system is likely to be responsible for the rapid


technology

285

growth of the e-recruitment jobseekers, where jobseekers only need to post theirresumes once to the e-recruiters’ web sites. For continual job application, the jobseekersonly click to accept the application to the company of interest without resending theresumes and personal information.

Behavioural intention (BI)According to Bagozzi et al. (1992) new technologies such as personal computers arecomplex and an element of uncertainty exists in the minds of decision makers withrespect to the successful adoption of them, people form attitudes and intentions towardtrying to learn to use the new technology prior to initiating efforts directed at using.Attitudes towards usage and intentions to use may be ill-formed or lacking inconviction or else may occur only after preliminary strivings to learn to use thetechnology evolve. Thus, actual usage may not be a direct or immediate consequence ofsuch attitudes and intentions.

Sanchez-Franco and Roldan (2005) study of Web acceptance among experientialusers and goal-directed users, on the relationship between PU and BI (H3) found that itwas not significant among the experiential users, thereby rejecting H3. According tothese authors, experiential users would not engage in an experiential and playfulbehaviour that also increases extrinsic rewards without previously adjusting theirattitudes. However, usefulness-influence on intention to use web among goal-directedusers is greater than among experiential, supporting H3a. Consequently, this studyrelates PEOU to PU and PU to BI with the following hypotheses:

H1. PEOU is positively related to PU in e-recruitment adoption.

H2. Perceived usefulness is positively related to behavioural intention to usee-recruitment for job search.

Perceived privacy risk (PPR)In the consumer behaviour literature and consumer decision-making process research,perceived risk (PR) concept is often considered. Since its introduction by Bauer (1960)and the introduction of internet, many IT researchers adopt this concept to study andunderstand the users’ evaluation on these PRs as “obstacles” to computer technologyadoption.

Most consumer behaviour literatures evaluated on PRs are monetary related exceptLiebermann and Stashevsky (2002) findings on the validity of personal informationstealing. This issue relates to jobseekers posting their resumes to the job sites. Forexample, there are reported cases where some head-hunters are able to “unlock”corporate web sites and roam the site for staff directories, resumes, photos, andorganizational charts, which later were on sale either to recruitment agencies, ordirectly to companies that may be interested (Galanaki, 2002).

In fact, recent poll in Business Week reviewed the importance of internet privacywas underscored. According to the March 16, 1998 issue, 61 per cent of those not onlinesaid they would use the internet if they felt their privacy (Introna and Pouloudi, 1999),particularly, the jobseekers’ resumes and applications could be protected (Guttermanet al., 1999). Galanaki (2002) also described that for e-recruitment businesses, the majorethical issue is the concern of confidentiality and trust during resume handling bye-recruiters. In short, given that PPR is inevitable for jobseekers posted resumes and

IMDS109,2

286

applications and the possibility of being hacked, reviewed by employers, and subject totempering by others. Therefore, the proposed hypothesis is as follow:

H3. PPR adversely affects PU in e-recruitment adoption.

Performance expectancy (PE)According to Compeau and Higgins (1995), outcome expectations exert a significantinfluence on individuals’ reactions to computing technology. Bandura (1986) explainedthat the expected consequences of one’s behaviour might construe as an influence onaffect (or liking) for the behaviour through a process of association. The satisfactionderived from the favourable consequences of the behaviour becomes linked to thebehaviour itself, causing an increased affect for the behaviour.

In the job search literature, Baik et al. (1989) study estimated correlates ofpsychological distress with a heterogeneous sample of American subjects (N ¼ 122;M age ¼ 33.5 year) who were involuntarily displaced from work. Their results showthat expectation of finding a new job is a significant source of psychological distress injob loss after taking into account length of unemployment and economic dependence.Psychological distress also showed significant negative association with self-esteemand significant positive relationship with job seeking effort.

Performance expectations are similar to the PU in TAM, where users tend toundertake behaviours they believe will help them perform their job better (Compeauand Higgins, 1995). However, they highlighted that the model tested is incomplete andno conclusion statements about causality since Social Cognitive Theory (SCT) is basedon a continuous reciprocal interaction among the factors studied, which they suggestedfeedback mechanism to be modelled on future study.

In view of this, the author sets to test Bandura’s first set of outcome expectation,that is, the expectation relates to outcome and introduces it as PE as an externalvariable to TAM. Jobseekers would view outcome expectation of e-recruitment useful ifit is more effective than other recruitment sources. In this respect, they would expectpositive PE as they might increase chances of being spotted by the e-recruiters’ clients,reaching them in time, and spending less time on repeated applications. Hence, theperformance expectation is expected to have direct influence on PU and behaviouralintention to use e-recruitment and it is hypothesized that:

H4. Performance expectation outcome is positively related to perceive usefulnessin e-recruitment adoption.

H4a. PE correlates with ASSE in e-recruitment adoption.

H4b. PE correlates with PPR in e-recruitment adoption.

Application specific self-efficacy (ASSE)Self-efficacy as defined by Bandura (1986) is the people’s judgement of theircapabilities to organize and execute courses of action required to attain designatedtypes of performances. It is concerned not with the skills one has but with judgementsof what one can do with whatever skills one possesses. The self-efficacy construct hasalso been included in many studies involving the TAM, including those of Yi andHwang (2003), Chau and Hu (2001), Igbaria et al. (1995), and Venkatesh and Davis(1996).


technology

287

Besides, Compeau and Higgins (1995) also allege that self-efficacy determinescomputer usage, both directly and through outcome expectations. They therefore,concluded that SCT perspective suggests that an understanding of two distinctdimensions of outcome expectation and self-efficacy is necessary to understandcomputing behaviour. Their findings indicated that self-efficacy influence usagedirectly, as well as indirectly, through outcome expectations, affect, and anxiety.Outcome expectations influence usage directly, as well as indirectly, through affect.

According to Marakas et al. (1998), CSE is a multi-level construct operating at twodistinct levels: at general computing level (general CSE) and at the specific applicationlevel application-specific self-efficacy (ASSE). General CSE is defined as an individualjudgement of efficacy across multiple computer domains and ASSE is defined as anindividual perception of self-efficacy in using a specific application or system withinthe domain of general computing.

The result of Yi and Hwang (2003) study also confirmed that ASSE had asignificant effect on ease of use (b ¼ 0.49, p , 0.001), supporting their hypothesis. Theauthors concluded that ASSE has been shown to exert a significant effect on systemuse over and above behavioural intention (BI). This confirms that both BI and ASSEare determinants of actual system use, a central dimension of technology acceptancebehaviour. Therefore, the author suggests that ASSE is one of the external variablesthat should be considered along with BI. Thus, the following hypotheses areestablished:

H5. Application-specific self-efficacy is positively related to PU in e-recruitmentadoption.

H5a. Application-specific self-efficacy correlates with PPR in e-recruitmentadoption.

Perceived stress (PS)In “internet Self-efficacy and the Psychology of the Digital Divide,” Eastin and LaRose(2000) defined stress encountered while using the internet is the number of stressorsencountered while online. Having trouble getting on the internet, the difficulty tocomplete the e-application forms, resume update reminder and computer freezes up arecommon examples. When jobseekers encounter such problems, it might lowerexpectation about successful interactions with the internet in the future. As the numberof stressors encountered online increase, perceptions of success decrease andself-efficacy along with it.

Moreover, Eastin and LaRose (2000) found that negatively related to internetself-efficacy is internet stress and Self-disparagement. This single study limits thevalidity of a construct and the convenience sample used restricts the generalizability ofthe results. Hence, they suggest the future research should investigate the interplayamong internet self-efficacy, stress and online support. Therefore, e-recruitmenttechnology system designed for user-friendly is crucial. With the perceptions of someform of control, jobseekers would expect an easy to use e-resume blank in thee-recruitment platform. The internet stress, in particular, the PS experienced bye-recruitment users has not been studied. With jobseekers’ perception on stress lesssystem; PS becomes ease of use that will motivate them to frequent utilization of thetechnology. Hence, the proposed hypotheses are as follow:

IMDS109,2

288

H6. PS is positively related to PEOU in e-recruitment adoption.

H6a. PS correlates with PPR in e-recruitment adoption.

H6b. PS correlates with application-specific self-efficacy in e-recruitment adoption.

H6c. PS correlates with PE in e-recruitment adoption.

Therefore, given this empirical tested study of modified TAM and its significant causallink among the three constructs by previous researchers, the author attempts to useStructural Equation Modelling (SEM) to test these highly validated studies with PEOU,PU, as independent variables and BI as the dependent variable for this study.

Extended to this modified TAM are four validated external variables that consist ofBandura’s SCT of Performance Expectation (PE) and Application specific self-efficacy(ASSE), Perceived Privacy Risk (PPR), and (PS). Since this study integrates fourdifferent theoretical systems of PE, ASSE, PPR, and PS to PU and PEOU asantecedents, the author proposes to examine the interrelationships among theseantecedent variables that are still unexplored in the modified TAM literature. Thus, theresearch framework is proposed in Figure 1.

MethodsProcedures and participantsTo test the aforementioned hypotheses, an empirical study was carried out. The authorinitially engaged five “part-timers” in January 2007 to obtain the data from employedjobseekers with third-party e-recruitment experience but the success rate was low andthe duration of collection exceeded the deadline set. Only 31 questionnaires werecollected. With this experience, the author weighed the additional cost and time againsthired cost and the probability of low response rate, thus opted for the SnowballSampling (Patton, 1990) strategy as non-probability sampling.

Figure 1.Research framework for

employed jobseekerse-recruitment technology

adoption

Perceived Privacy Risk H3

H4b

H4H5a

Outcome Expectation

H4aH6a

H5

H6c Computer Self-efficacy H6b

H6

Internet Stress

H2

H1

Modified Technology Acceptance Model

PerceivedUsefulness (PU)

Perceived Ease ofUse (PEOU)

Behavioural Intentions toUse (BI)

Perceived PrivacyRisk (PPR)

Application specificself-efficacy

(ASSE)

PerceivedStress (PS)

PerformanceExpectancy (PE)


technology

289

Although this method would hardly lead to representative samples, there are timeswhen it may be the best method available. Snowball sampling is especially useful whenyou are trying to reach populations that are inaccessible or hard to find (Trochim, 2002)in particular, the employed jobseekers who might be keeping a “low profile” of usinge-recruitment for job search. Using this technique, the author distributed the surveyquestionnaires a month later in soft-and-hard copies by e-mail and mail, respectively,to friends, relatives, and colleagues who met the criteria of being employed in anyorganization with experienced in third-party e-recruitment usage. Likewise, this groupin turn distributed the questionnaires in this manner to their friends, relatives, andcolleagues, thus, “snowballing” to obtain sufficient participants.

The respondents were asked to complete a six-page questionnaire that consisted of31 items. All items were measured on a 5-point Likert type scale and respondents wereasked to indicate their perceptions and experiences of the e-recruitment usage on eachitem ranging from “1 ¼ strongly disagree” to “5 ¼ strongly agree.” A total of 283respondents replied in which 20 percent of them by e-mails and 80 percent by handsand mails. After sorting those questionnaires with missing data, 262 sets were valid foranalysis.

The 262 participants in this study consists of 136 male (51.1 percent) and 130 female(48.9) with the age group (SD ¼ 1.420) and the highest groups were 21-25 (24.8 percent),26-30 (33.8 percent), 31-35 (19.5 percent), 36-40 (12.0 per cent), 41-45 (3.8 percent),46-50 (3.8 percent), and above 50 (1.5 percent). The qualifications varied betweensecondary/high school to post graduates and higher but the highest percentage being theuniversity (Bachelor Degree) category, with 133 (50.0 per cent) respondents.

ResultsEstimation of measurement modelUsing Statistical Package for the Social Science (SPSS) software version 12.0, thedescriptive statistics and principal components exploratory factor analyses withvarimax rotation were conducted. For the purpose of confirmatory factor analyses andthe relationships between the constructs the conceptual framework was tested usingAnalysis of Moment Structures (AMOS) Version 4.01 (Arbuckle, 1997).

The evaluation process began by initially performing all the 31 observed variablesat univariate level for normality. Examining the skew and kurtosis estimates allobserved measures were less than the absolute value of three in terms of skew, and lessthan the absolute value of eight in term of kurtosis. Therefore, the univariatedistributions looked reasonably symmetric. For multicollinearity test, the Pearsoncorrelation matrix was performed. The matrix shows that none of the coefficients isgreater than 0.8, indicating that it can be judged as no significant violation to thenon-multicollinearity assumption Garson (2004, May).

Reliability analysisThe 31-items for employed jobseekers’ e-recruitment experience and perception, whichcomposed of 5-item PU scale, 5-item PEOU scale, 4-item PPR scale, 4-item ASSE scale,4-item perceived stress (PS) scale, 4-item PE scale, and 5-item BI scale were tested andthe internal consistency reliabilities are all above 0.7, which was intended as theminimum cut-off alpha measure (Cronbach, 1951). All the constructs met the internalconsistency reliabilities with the lowest measure of 0.74 and highest 0.83.

IMDS109,2

290

Validity analysisThe validity of the scales was verified by considering the content validity, convergentvalidity, and discriminant validity (Hair et al., 1998). The convergent validity conceptevaluates the extent to which two measurements of the concept may be correlated andusing the Variance Extracted Calculation the computation shows that all the constructsmeet the threshold value of 0.50. The content validity is confirmed following anextensive review of the job search and e-recruitment literatures.

The discriminant validity refers is the degree that measures of different constructsare internally correlated, distinct from other constructs and unique and can be assessedby square root of the average variance extracted (AVE) between the constructs andtheir measures Hair et al. (1998). If the squared correlation coefficients (R2) are lowerthan AVE, the constructs have discriminant validity. Based on this calculation, theconstructs were found to be larger than the squared correlation coefficients (R2).Table II also shows the mean, SD, internal reliabilities, correlations, and AVE of theconstructs. Thus, the instrument relatively passes in the three tests, which suggestsstrong convergent validity for the research variables.

Estimation of proposed causal modelThe final approach to model assessment is to compare the proposed model with aseries of competing models, which act as alternate explanations to the proposed model.In this way, the author can determine whether the proposed model, regardless ofoverall fit (within reasonable limits), is acceptable because no other similarlyformulated model can achieve a higher level of fit. This step is particularly importantwhen the chi-square statistic indicates no significant differences in overall model fitbecause there may always be a better-fitting model, even in the case of non significantdifferences (Hair et al., 1998).

For the above purpose, Garson (2004 May) suggests by initially over fit theproposed model, then changing only one parameter at a time to obtain theparsimonious model with references to Chi-square ratio (x 2/df), NFI, TLI, CFI, rootmean square (RMSEA), goodness-of-fit (GFI), and AGFI.

The initial results indicated the indices were within the acceptable level but both PSand PEOU measurement error terms were greater than one. Both PS and PEOUconstructs were eliminated in turn. The final test-retest of the competing modelsrequired a total deletion of PS construct to achieve a parsimonious model with

Mean SD ICR 1 2 3 4 5 6

1. PE 2.96 1.17 0.80 0.732. ASSE 2.88 1.13 0.79 0.37 0.583. PPR 3.02 1.20 0.83 0.29 0.40 0.814. PEOU 3.39 1.21 0.75 0.00 0.086 0.10 0.635. PU 2.35 1.15 0.81 0.34 0.45 0.58 0.10 0.716. BI 2.49 1.04 0.74 0.23 0.27 0.36 0.06 0.33 0.73

Notes: ICR ¼ Internal consistency reliabilities. Diagonal elements in italics are the square root ofAVE between the constructs and their measures. Off-diagonal elements are correlations betweenconstructs. For discriminant validity, the diagonal elements should be larger than off-diagonalelements in the same row and column (Yi and Hwang, 2003)

Table II.Internal consistencies,

correlations, and AVE ofconstructs


technology

291

Chi-square ratio (2.103), NFI (0.908), TLI (0.934), CFI (0.949), RMSEA (0.065), GFI(0.911), AGFI (0.870), and 90 per cent Confidence Interval (0.053, 0.076), which showsgood psychometric properties.

Based on the output, the final set of PEOU construct with two components, PEOU 2and PEOU 4 to PU was found to have low standardized regression weight. Therefore,from the causal perspective, the regression results suggest that PEOU is a weakantecedent to PU for e-recruitment adoption. The only moderately strong componentrelationship is PEOU2 to PU as expressed by employed jobseekers are the flexibility tointeract with the e-recruitment system with squared multiple correlations of R 2 of0.597, which account for only 59.5 percent of the observed variance shown in Figure 2.

The PEOU with standardized regression weight of 0.072 was found to be nonsignificant path to PU. This means hypothesis H1 is not supported. In TAM literature,the original TAM model to have a stronger support of PEOU with PU and mayactually be a prime causal antecedent of PU (Davis et al., 1989). Many past studies havealso demonstrated that PEOU to PU was significant and positively related(Sanchez-Franco and Roldan, 2005; Ma and Liu, 2004; Featherman and Pavlou, 2003;Teo, 2001). However, other studies have dissimilar findings, indicating PEOU and PUrelationship is inconsistent and weak. For example, the study by Morris and Dillon(1997) study showed its weak relationships on the influence of user perceptions onsoftware utilization and (Klopping and McKinney, 2004; Yi and Hwang, 2003; Chau andHu, 2001; Venkatesh, 1999) the inconsistence to PU and attitude formation. This studycomplies with the latter group of findings.

The path linear relationship between PU and BI is considerably strong withstandardized regression weight 0.610, thus supporting H2. However, the squaredmultiple correlation (R 2) for BI was 0.372, which account for only 37.2 percent of theobserved variance while explaining the intention to use e-recruitment. The path linearrelationship between PU and BI suggests that PU has a direct effect on BI and is acritical factor for employed jobseekers’ acceptance of e-recruitment technology.This finding is also consistent with the several recent TAM studies that suggest PUis more important than PEOU in determining whether or not to use a technology(Fusilier and Durlabhji, 2005; Venkatesh, 2000; Chau, 1996; Igbaria et al., 1996).

The path PPR and PU is significant with standardized regression weight of 0.459.This indicates that hypothesis H3 is supported. In the PR literature, notably privacy

Figure 2.Final parsimonious modeland its related hypotheses

H4b = 0.45* H3 = 0.459*

H5a = 0.77* H2 = 0.610*

H4a = 0.75*H1 = 0.072 (n.s)

H5 = 0.591*

Standardized path coefficient for Employed Jobseekers

Notes: ns: non-significant; *p < 0.05; **p < 0.01; ***p < 0.001

PE

ASSE

PPR

PEOU

BIPU

IMDS109,2

292

and security, privacy attracts considerable attention as increasing amounts ofinformation flow through various electronic communication channels (Featherman andPavlou, 2003; Introna and Pouloudi, 1999). Similarly, employed jobseekers have thesame concerned on privacy.

The above hypothesis H4 is not supported. The proposed PE causal path to PU wasdeleted to obtain the final model fit indices. This means PE does not have direct effecton PU to use e-recruitment. The hypothesis H5 is supported with path coefficient 0.591,indicating that ASSE is positively related to PU in e-recruitment adoption. The totaldeletion of PS construct means H6. The summary of the main and sub-hypotheses isshown in Table III.

Discussion and implicationsNowadays, it is a known fact that using internet for recruitment is an upward trend.This study has empirical reviewed the findings of the employed jobseekers’perceptions and experiences on e-recruitment adoption for job search. Based on thefindings of the final model, this paper has identified few key indicators to e-recruitmentadoption, thus contributing to the existing knowledge in the human resourcesliterature, particularly in recruitment.

The PEOU construct indicates that the employed jobseekers could comprehend andbecome familiar with the operation of e-recruitment technology quickly over time.

Hypothesis SupportedPath coefficient

(PC)/correlation (C)

H1: Perceived ease of use is positively related toperceived usefulness in e-recruitment adoption No 0.072 * (PC)

H2: Perceived usefulness is positively related to BI to usee-recruitment for job search Yes 0.610 * (PC)

H3: Perceived privacy risk adversely affects perceivedusefulness in e-recruitment adoption Yes 0.459 * (PC)

H4: Performance expectation is positively related toperceive usefulness in e-recruitment adoption No No

H4a: Performance expectancy correlates with ASSE ine-recruitment adoption Yes 0.75 * (C)

H4b: Performance expectancy correlates with perceivedprivacy risk in e-recruitment adoption Yes 0.45 * (C)

H5: Application-specific self-efficacy is positively relatedto perceived usefulness in e-recruitment adoption Yes 0.591 * (PC)

H5a: Application-specific self-efficacy correlates withperceived privacy risk in e-recruitment adoption Yes 0.77 * (C)

H6: Perceived stress is positively related to perceivedease of use in e-recruitment adoption Construct deleted

H6a: Perceived stress correlates with perceived privacyrisk in e-recruitment adoption Construct deleted

H6b: Perceived stress correlates with application-specificself-efficacy in e-recruitment adoption Construct deleted

H6c: Perceived stress correlates with performanceexpectancy in e-recruitment adoption Construct deleted

Notes: *p , 0.05; * *p , 0.01; * * *p , 0.001

Table III.The hypotheses results

for extended TAM


technology

293

The e-application blanks/forms used in most e-recruitment web sites generally aredesigned similar to the traditional pen-and-paper job application blank formats, whichare easy to follow. This suggests e-recruiters should maintain an easy to usee-application blank format for jobseekers.

However, employed jobseekers perceived usefulness (PU) in e-recruitmenttechnology is more important and it indicates that detail job information would leadthem to better decisions. A good decision is a logical one based on the availableinformation and reflecting the preferences of the decision maker (Harris, 1998).Jobseekers expected in-depth job information such as, job descriptions, quickresponses, and links to corporate web sites (Gowan, 2001). The Signalling Theory alsosuggests that prospective employees who receive comprehensive information fromcompanies during the recruitment process regarding the companies’ goals, cultures,and general philosophies of doing business will tend to be more self-selective toapplying to companies that best fit their personal goals and philosophies (Brice &Waung, 2002).

The low percentage of observed variance for BI to use e-recruitment could probablysuggest that employed jobseekers do not use e-recruitment technology for job searchper se but probably still combined with other conventional job search methods. A studyof college students found that there were significant correlation between the number ofjob search methods used and the number of interviews and suggested students to use avariety of job search methods rather than rely on a single method (Mau and Kopischke,2001). This implies that employed jobseekers intended to seek for better jobopportunity will use multiple methods to obtain the jobs of interest. Therefore, humanresource managers should consider using multiple recruitment methods only forurgent recruitment as it affects increase of recruitment cost per hire.

This study also implies that employed jobseekers preference on using e-recruitmentis to survey for job market value. If there is a prospective job opportunity they wouldapply for the job. The job market value search is a process of determining competitivepay levels for specific jobs in a defined external market. In assessing the job marketvalue, jobseekers usually read employment advertisements, which often indicate thepackages on offer. This will also ensure they keep up to date with the skillsrequirements and current industry trends as advertised by the employers(MyCareer.com, 2005). Presumably, this also explains why some employedjobseekers remain passive and would delay applying for jobs if external job marketvalue proves little or no significant different from their current job prospects.

Vallerand and Bissonnette (1992) explained that for intrinsic motivation to occur, atask must offer some opportunity that meets a person’s inner needs. In these cases whereintrinsic and extrinsic motivations are mixed, one might suppose that over a period oftime the accompanying extrinsic reinforcements gradually increase our intrinsicenjoyment of the activity. . . and perhaps vice versa. That is, a high salary may, in time,make the work seem more enjoyable (Tucker-Ladd, 2000). Based on the result obtained,employed jobseekers are intrinsically motivated and willing to recommend others to usethe same e-recruitment web site if it has helped them to obtain jobs. This implies that onemethod to spread the e-recruitment technology fast would be by word-of-mouth amongthe users who have been successfully recruited adopting this technology.

The elimination of the proposed PE causal path to PU suggests that the jobseekersdo have high expectation of being spotted by e-recruiters when applying for jobs.

IMDS109,2

294

However, not all jobseekers are lucky and may face very slim chance of being spotted,especially in the renowned e-recruitment web sites with millions of resumes (Cappelli,2001). Therefore, the chances of being spotted and short-listed, obtaining an interviewand job would be lower compared to other job search methods. A proposedimprovement to this is to incorporate a “powerful” matching system that could assistthe clients to sort their advertised job requirements to job applications so that talentedjobseekers would not be overlooked.

The ASSE positive linear relationship to PU exert a direct significant effect on BI touse e-recruitment suggests the importance communication as it might affect applicant’sreaction to the recruitment and selection processes. Time is crucial for both jobseekersand recruiters, especially when the nature of jobs is changing faster than people;this suggests that when employed jobseekers require additional job-related informationthey would expect real-time communication and responses from the e-recruiters(Burton Cober et al., 2000). In addition, e-recruiters should also inform the jobseekerse-applications status. Failing which it would deter employed jobseekers adoption as itmakes no difference from other conventional methods of application.

The adoption of this technology has caused some concerns on privacy. Although,Useem (1999 July) has cautioned that some employers hired full-time “salvagers” topatrol and view the resume posted in the e-recruitment sites, jobseekers are aware ofthe risks and their continual usage suggests that the chances of being spotted by theemployers among many posted resumes would be slimmed. Another explanation isthat internet cannot provide security in probably the next five to ten years to come(Bryant, 2000). Overall, the employed jobseekers perceiving these risks but stilladopting e-recruitment suggest the risk is worth taking because of its ease of use,usefulness, application posting speed, and advantageous over other job applicationmethods.

The total elimination of PS construct in the final model implies that althoughinternet technology requires some computer skills for adoption, these employedjobseekers with working experiences may not find usage of internet technologystressful due to daily usage and operation of computers. This explains that if peoplebelieve they can exert some control over stressors, they usually have less impact(Bernstein et al., 1997; Nairne, 2000). Therefore, variation of PS dimension is not aconcern to employed jobseekers for e-recruitment adoption.

LimitationThere is few limitations concern the generalisability of the findings in which thesample was collected by snowball sampling and has a predominance of youngerrespondents with degree qualifications and study was conducted in Malaysia setting,and uses self-report scales, which might inflate correlations through common methodsvariance (Fusilier and Durlabhji, 2005). However, it provides some indications of theemployed jobseekers’ intent of e-recruitment adoption, which can be replicated in othercountries using the same model and instrument to identify and consolidate employedjobseekers’ perceptions and behaviour toward this technology adoption.

Conclusion and recommendationThis research finding implies two important factors for third party e-recruiters’ policymakers and human resource management practices. It is evident that the e-recruitment


technology

295

system and services need further improvement to recruit employed jobseekersspecifically the passive talented candidates who used e-recruitment platform for jobmarket value survey. The system improvement should incorporate by:

. finding the right method in this technology to attract them in the initial stage ofrecruitment process; and

. finding the right system to auto-match the candidates’ knowledge, skill, andabilities with the job requirements in the initial stage of selection processprobably by electronic sifting.

Lievens and Highhouse (2003) have similar view on applicants’ attraction by usingmarketing-based angle to the attractiveness of organizations in the early stages of therecruitment process. They believed that potential applicants’ initial attraction to anorganization, as a place to work cannot be explained solely on the basis of job andorganizational factors. Applicants’ initial attraction to an organization employingjobseekers is also based on the symbolic meanings (in terms of inferred traits) that theyassociate with organizations and will use these trait inferences as points ofdifferentiation among various employing organizations. The lack of research data onattraction outcomes can be attributed to a general lack of recruitment assessment(Connerley et al., 2003). Hence, the conceptualization of brand image in the marketingliterature can be applied in future e-recruitment study.

Most of the third-party e-recruitment web sites have incorporated some kind ofelectronic sifting systems but generally depend on the jobseekers’ input of theirqualifications and their expertise in the database. However, using e-sifting may be veryfast, but not necessary any more accurate (Cook, 1998). Future improvement of thise-shifting system should consider this issue to make the system more effective.Feedback on the accuracy of the system from jobseekers and clients is recommended(Liu and Wang, 2007; Karlsen et al, 2006).

E-recruiters should also consider continuing improvement on the web site privacyprotection, particularly shielding their members’ personal information and resumes frombeing viewed by their employers. This improvement is possible if the system is designedto allow jobseekers to bar their own company from viewing their applications.

With the recent rising costs in third-party e-recruitment, many corporate companiesare now incorporating their web sites for e-recruitment (Harrington, 2002; Gordon,2002). Therefore, further study should compare the effectiveness the third-party andcorporate companies’ e-recruitment methods to recruit talent employees with genderand race as moderating variables. The study of the threats among non-profitorganizations, corporate companies, newspapers, and executive search e-recruitment tothird-party e-recruiters is also recommended.

References

Agarwal, R., Sambamurthy, V. and Stair, R.M. (2000), “Research report: the evolving relationshipbetween general and specific computer self-efficacy – an empirical assessment”,Information Systems Research, Vol. 11 No. 4, pp. 418-30.

Arbuckle, J. (1997), AMOS User’s Guide, Smallwaters, Chicago, IL.

Bagozzi, R.P., Davis, F.D. and Warshaw, P.R. (1992), “Development and test of a theory oftechnological learning and usage”, Human Relations, Vol. 45 No. 7, pp. 660-86.

IMDS109,2

296

Baik, K., Hosseini, M. and Priesmeyer, H.R. (1989), “Correlates of psychological distress ininvoluntary job loss”, Psychological Reports, Vol. 65 No. 3, pp. 1227-33.

Bandura, A. (1986), Social Foundations of Thought and Action: A Social Cognitive Theory,Prentice-Hall, Englewood Cliffs, NJ.

Bauer, R. (1960), “Consumer behaviour as risk taking”, in Hancock, R.F. (Ed.), Proceedings of the43rd Conference of the American Marketing Association, American MarketingAssociation, Chicago, IL, pp. 389-98.

Bernstein, D.A., Clarke-Stewart, A., Roy, E.J. and Wickens, C.D. (1997), Psychology, 4th ed.,Houghton Mifflin Company, Boston, MA.

Brice, T.S. and Waung, M. (2002), “Web site recruitment characteristics: America’s best versusAmerica’s biggest”, SAM Advanced Management Journal, Vol. 67 No. 2, pp. 4-8.

Bryant, S. (2000), “Top executives shun internet recruitment”, Management Accounting:Magazine for Chartered Management Accountants, Vol. 78 No. 4, p. 1/4.

Byars, L.L. and Rue, L. (2000),HumanResourceManagement, 6th ed., McGraw-Hill, New York, NY.

Cappelli, P. (2001), “Making the most of on-line recruiting”, Harvard Business Review, Vol. 79No. 3, pp. 139-46.

Cascio, W.F. (1998), Applied Psychology in HRM, Prentice-Hall, London.

Chau, P.Y.K. (1996), “An empirical assessment of a modified technology acceptance model”,Journal of Management Information Systems, Vol. 13 No. 2, pp. 185-204.

Chau, P.Y.K. and Hu, P.J. (2001), “Information technology acceptance by individual professionals:a model comparison approach”, Decision Sciences, Vol. 32 No. 4, pp. 699-719.

Cober, R.T., Brown, D.J., Blumental, A.J., Doverspike, D. and Levy, P. (2000), “The quest for thequalified job surfer: it’s time the public sector catches the wave”, Public PersonnelManagement, Vol. 29 No. 4.

Compeau, D.R. and Higgins, C.A. (1995), “Computer self-efficacy: development of a measure andinitial test”, MIS Quarterly, Vol. 19 No. 2, pp. 189-211.

Connerley, M.L., Carlson, K.D. and Mecham, R.L. III (2003), “Evidence of differences in applicantpool quality”, Personnel Review, Vol. 32 No. 2, pp. 22-39.

Cook, M. (1998), Personnel Selection: Adding Value through People, 3rd ed., Wiley, New York, NY.

Cronbach, L.J. (1951), “Coefficient alpha and the internal structure of tests”, Psychometrika,Vol. 16, pp. 297-334.

Davis, F.D. (1993), “User acceptance of information technology: system characteristics, userperceptions and behavioural impacts”, International Journal of Man-Machine Studies,Vol. 38, pp. 475-87.

Davis, F.D., Bagozzi, R.P. and Warshaw, P.R. (1989), “User acceptance of computer technology:a comparison of two theoretical models”, Management Science, Vol. 35 No. 8, pp. 982-1003.

Davis, F.D., Bagozzi, R.P. and Warshaw, P.R. (1992), “Extrinsic and intrinsic motivation to usecomputers in the workplace”, Journal of Applied Social Psychology, Vol. 22, pp. 1111-32.

Dixon, P. (2000), Job Searching Online for Dummies, IDG Books Worldwide Inc., Boston, MA.

Eastin, M.S. and LaRose, R. (2000), “Internet self-efficacy and the psychology of the digitaldivide”, Journal of Computer Mediated Communication, available at: www.ascusc.org/jcmc/vol6/issue1/eastin.html (accessed 29 November 2000).

Featherman, M.S. and Pavlou, P.A. (2003), “Predicting e-services adoption: a perceived risk facetsperspective”, International Journal of Human-Computer Studies, Vol. 59, pp. 451-74.

Fisher, S. (2001), “The road to profitability”, InfoWorld, Vol. 23 No. 28, p. 32.


technology

297

Fountain, C. (2005), “Finding a job in the internet age”, Social Forces, Vol. 83 No. 3, pp. 1235-62.

Fusilier, M. and Durlabhji, S. (2005), “An exploration of student internet use in india”,Campus-Wide Information Systems, Vol. 22 No. 4, pp. 233-46.

Galanaki, E. (2002), “The decision to recruit online: a descriptive study”, Career DevelopmentInternational, Vol. 7 No. 4, pp. 243-51.

Garson, D.G. (2004), “Structural equation modelling”, available at: www2.chass.ncsu.edu/garson/pa765/structur.htm (accessed 20 May 2004).

Gomolski, B. (2000), “E-business matters: the five success factors needed to survive the dot-comshakeout (industry trend or event)”, InfoWorld, Vol. 22 No. 47, p. 102.

Gordon, J. (2002), “Battle of the boards”, Forbes, Vol. 70 No. 3, p. 50.

Gowan, M. (2001), “Find the right job online (industry trend or event)”, PC World, Vol. 19 No. 2,p. 135.

Gutterman, A.S., Brown, R. and Stanislaw, J. (1999), The Professional’s Guide to Doing Businesson the Internet, Harcourt Brace Professional Publishing, San Diego, CA.

Hair, J.F. Jr, Anderson, R.E., Tatham, R.L. and Black, W.C. (1998), Multivariate Data Analysis,Prentice-Hall, Englewood Cliffs, NJ.

Harrington, A. (2002), “Can anyone build a better monster”, Fortune, Career Section, Vol. 145No. 10, pp. 189-92.

Harris, R. (1998), Introduction to Decision Making, available at: http://virtualsalt.com/crebook5.htm (accessed 12 April 2003).

Igbaria, M., Iivari, J. and Maragahh, H. (1995), “Why do individuals use computer technology?A Finnish case study”, Information & Management, Vol. 29, pp. 227-38.

Igbaria, M., Parasuraman, S. and Baroudi, J.J. (1996), “A motivational model of microcomputerusage”, Journal of Management Information Systems, Vol. 13 No. 1, pp. 127-43.

Introna, L.D. and Pouloudi, A. (1999), “Privacy in the information age: stakeholders, interest andvalues”, Journal of Business Ethics, Vol. 22 No. 1, pp. 27-38.

Joyce, E. (2002), Bill Warren, E-recruiting Association, available at: http://atnewyork.com/news/article.php/1433771 (accessed 26 July 2002).

Karlsen, J.T, Anderson, J., Birkley, L.S. and Odegard, E. (2006), “An empirical study of criticalsuccess factors in IT projects”, International Journal of Management and EnterpriseDevelopment, Vol. 3 No. 4, pp. 297-311.

Klopping, I.M. and McKinney, E. (2004), “Extending the technology acceptance model and thetask-technology fit model to consumer e-commerce”, Information Technology, Learning,and Performance Journal, Vol. 22 No. 1.

Kuhn, P. and Skuterud, M. (2000), “Job search methods: internet versus traditional”, MonthlyLabour Review, Vol. 123 No. 10, pp. 3-11.

Labanyi, D. (2002), “E-recruitment services continue to prosper as more people go online to find ajob”, The Irish Times, 20 December, p. 51.

Liebermann, Y. and Stashevsky, S. (2002), “Perceived risks as barriers to internet ande-commerce usage”, Qualitative Market Research: An International Journal, Vol. 5 No. 2,pp. 291-300.

Lievens, F. and Highhouse, S. (2003), “The relation of instrumental and symbolic attributes to acompany’s attractiveness as an employer”, Personnel Psychology, Vol. 56 No. 1, p. 102.

Liu, C.C. and Wang, H.J. (2007), “Developing measures of digital capital and virtual value chainconstruction in job search websites”, International Journal of Management & EnterpriseDevelopment, Vol. 4 No. 1, pp. 66-81.

IMDS109,2

298

Lopez, D.A. and Manson, D.P. (1997), “A study of individual computer self-efficacy and perceivedusefulness of the empowered desktop information system”, Journal of InterdisciplinaryStudy, Fall, available at: www.csupomona.edu/ , jis/1997/Lopez.pdf (accessed 13 June2005).

Ma, Q. and Liu, L. (2004), “The technology acceptance model: a meta-analysis of empiricalfindings”, Journal of Organizational and End User Computing, Vol. 16 No. 1, pp. 59-72.

Malhotra, Y. and Gelletta, D.F. (1999), “Extending the technology acceptance model to accountfor social influence: theoretical bases and empirical validation”, Proceedings of the 32ndHawaii International Conference on System Sciences, Maui, Hawau, 5-8 Januvary.

Marakas, G., Yi, M. and Johnson, R. (1998), “The multilevel and multifaceted character ofcomputer of computer self-efficacy: toward clarification of the construct and an integrativeframework for research”, Information System, Vol. 9 No. 2, pp. 126-63.

Mathieson, K. (2001), “Predicting user intentions: comparing the technology acceptance modelwith the theory of planned behaviour”, Information Systems Research, Vol. 2 No. 3,The Institute of Management Sciences, pp. 173-91.

Mau, W.C. and Kopischke, A. (2001), “Job search methods, job search outcomes and jobsatisfaction of college graduates: a comparison of race and sex”, Journal of EmploymentCounselling, Vol. 38 No. 3, p. 41.

Meisenheimer, J.R. II and Ilg, R.E. (2000), “Looking for a better job: job-search activity of theemployed”, Monthly Labour Review, Vol. 123 No. 9, p. 3.

Morris, M.G. and Dillion, A. (1997), “The influence of user perceptions on software utilization:application and evaluation of a theoretical model of technology acceptance”, IEEESoftware, Vol. 14 No. 4, pp. 58-65.

MyCareer.com (2005), Negotiating a Better Deal, available at: http://mycareer.com.au/advice-research/negotiating-a-better-deal.html (accessed 28 January 2005).

Nairne, J.S. (2000), Psychology: The Adaptive Mind, 2nd ed., Wadsworth, Belmont, CA.

Pastore, M. (2002), Global Online Population, available at: http://isp-planet.com/research/2002/census.html (accessed 28 January 2002).

Patton, M. (1990), Qualitative Evaluation and Research Methods, Sage, Newbury Park, CA.

Peters, K. (2001), “Five keys to effective e-cruiting”, Ivey Business Journal, Vol. 65 No. 3, pp. 8-10.

Pollitt, D. (2004), “Helps Xerox to pick the cream of the crop”, Human Resource ManagementInternational Digest, Vol. 12 No. 5, p. 33.

Pollitt, D. (2005), “Gets the Nike tick for approval”, Human Resource Management InternationalDigest, Vol. 13 No. 2, p. 33.

Quint, E.D. and Kopelman, R.E. (1995), “The effects of job search behaviour and vocationalself-concept crystallization on job”, Journal of Employment Counselling, Vol. 32 No. 2,pp. 88-96.

Rosencrantz, M.J. (1999), “The online job hunt”, CPA Journal, Vol. 69 No. 11, p. 15.

Said, S.A.G. and King, M. (1999), “Attitudes, satisfaction and usage: factors contributing to eachin the acceptance of information technology”, Behavioural and Information Technology,Vol. 18 No. 4, pp. 277-97.

Sanchez-Franco, M.J. and Roldan, J.L. (2005), “Web acceptance and usage model: a comparisonbetween goal-directed and experiential web users”, Internet Research, Vol. 15 No. 1,pp. 21-48.


technology

299

Teo, T.S.H. (2001), “Demographic and motivation variables associated with internet usageactivities”, Internet Research: Electronic Networking Applications and Policy, Vol. 11 No. 2,pp. 125-37.

Tong, D.Y.K. and Sivanand, C.N. (2005), “Service providers review: international andMalaysian”, Employee Relations, Vol. 27 No. 1, pp. 103-17.

Trochim, W.M.K. (2005), Probability and Non-probability Sampling, available at: http://socialresearchmethods.net/kb/sampprob.htm (accessed 20 March 2005).

Tucker-Ladd, C.E. (2002), Psychology Self-help, Mental Health Net, available at: http://mentalhelp.net/psyhelp/(accessed 28 January 2002).

Tyson, S. and York, A. (2000), Essentials of HRM, 4th ed., Butterworth-Heinemann, Oxford.

Useem, J. (1999), “Risk of posting resume”, Fortune, Vol. 139 No. 10, p. 290.

Vallerand, R.J. and Bissonnette, R. (1992), “Intrinsic, extrinsic, and motivational styles aspredictors of behaviour: a prospective study”, Journal of Personality, Vol. 60, pp. 599-619.

Venkatesh, V. (1999), “Creation of favourable user perceptions: exploring the role of intrinsicmotivation”, MIS Quarterly, Vol. 23 No. 2, pp. 239-60.

Venkatesh, V. (2000), “Determinants of perceived ease of use: integrating perceived behaviouralcontrol, computer anxiety and enjoyment into the technology acceptance model”,Information Systems, Vol. 11, pp. 342-65.

Venkatesh, V. and Davis, F.D. (1996), “A model of the antecedents of perceived ease of use:development and test”, Decision Sciences, Vol. 27 No. 3, pp. 451-81.

Yi, M.Y. and Hwang, Y. (2003), “Predicting the use of web-based information systems:self-efficacy, enjoyment, learning, goal orientation, and the technology acceptance model”,International Journal of Human-Computer Studies, Vol. 59, pp. 431-49.

Further reading

Blaikie, N. (2003), Analyzing Quantitative Data from Description to Explanation, Sage, London.

Karlsen, J.T., Andersen, J., Birkely, L.S. and Odegard, E. (2006), “An empirical study of criticalsuccess factors in IT projects”, International Journal of Management & EnterpriseDevelopment, Vol. 3 No. 4, pp. 297-331.

McKenna, E. and Beech, N. (1995), The Essence of Human Resource Management, Prentice-Hall,Englewood Cliffs, NJ.

Page, B. (1995), Learner’s Guide to the Memory Jogger II, GOAL/QPC, Salem, NH.

Peng, K.F., Fan, Y.W. and Hsu, T.A. (2004), “Proposing the content perception theory for theonline content industry – a structural equation modelling”, Industrial Management & DataSystems, Vol. 104 No. 6, pp. 469-89.

Venkatesh, V. and Davis, F.D. (2000), “A theoretical extension of the technology acceptancemodel: four longitudinal field studies”, Management Science, Vol. 45 No. 2, pp. 186-204.

Corresponding authorDavid Yoon Kin Tong can be contacted at: [email protected]

IMDS109,2

300

To purchase reprints of this article please e-mail: [email protected] visit our web site for further details: www.emeraldinsight.com/reprints

Documents

A Study of E-recruitment Technology Adoption in Malaysia