Staffing in the 21st Century Notes

Staffing in the 21st Century: Challenges and Opportunities

Staffing involves attracting, selecting, and retaining competent individuals to achieve organizational goals.
It's the primary way an organization shapes its diversity and human capital.
Modern organizations face staffing challenges like:
- Increased knowledge work demanding higher employee competencies.
- Demographic, labor, and cultural changes creating global talent shortages.
- Increasing workforce diversity.
A survey of 33,000 employers across 23 countries revealed that 40% struggled to find and hire desired talent (Manpower Inc., 2006).
Approximately 90% of nearly 7,000 managers indicated that talent acquisition and retention were becoming more difficult (Axelrod, Handfield-Jones, & Welsh, 2001).
Talent is rare, valuable, difficult to imitate, and hard to substitute (Barney & Wright, 1998).
Organizations that excel at attracting, selecting, and retaining talent are more likely to outperform competitors.
A study by Rynes, Brown, and Colbert (2002) revealed a significant gap between academic findings and managers' beliefs in the staffing domain.
Organizational decision-makers often fail to understand staffing or use it optimally, despite its strategic importance.
Organizations and scholars must recognize the value of staffing in the face of the global "war for talent."
The review provides a selective summary of key developments in staffing research since 2000.
It critically analyzes recruitment and personnel selection practices, offering suggestions for future theory and practical application.
It aims to evaluate the link between staffing theories/practices and organizational/business unit effectiveness.
There are significant gaps between research and practice, with limited research demonstrating the business value of staffing.
The review advocates for multi-level staffing research and models to bridge these gaps.
This review is limited to research published since roughly 2000, focusing on themes rather than chronology.
It considers how staffing contributes to outcomes at multiple levels of analysis, identifying gaps in understanding staffing research and practice.
The review aims to generate recognition of staffing's real and important impact on organizational effectiveness, while acknowledging the need to move beyond individual-level theories and methods.

Recruitment

Recruitment is defined as the organization's collective efforts to identify, attract, and influence job choices of competent applicants.
Organizational leaders recognize recruiting talent as a pressing problem.
Tight labor markets give applicants choices, especially in professional, knowledge-based, technical, and service occupations.
Nearly half of all employees are passively looking for jobs, and a sizable minority are actively searching (Towers Perrin, 2006).
Organizations face a greater recruiting challenge than a selection challenge.
Many organizations struggle with attracting a diverse workforce.
Recruiting is critical for sustained competitive advantage and organizational survival (Taylor & Collins, 2000).
Several reviews on recruitment exist (Breaugh & Starke, 2000; Highhouse & Hoffman, 2001; Rynes & Cable, 2003; Saks, 2005; Taylor & Collins, 2000).
Focus is on recent developments with implications for organizational effectiveness.
The Chapman et al. (2005) meta-analysis summarized 71 studies on recruiting predictors and applicant attraction outcomes.
Key findings from meta-analysis:
- Perceptions of person-organization fit (PO fit) and job/organizational attributes were the strongest predictors of recruiting outcomes.
- Perceptions of the recruitment process (e.g., fairness), recruiter competencies, and hiring expectancies were also important predictors.
- Recruiter demographics or functional occupation showed almost no relationship to the recruitment outcomes.
- Gender and study context (lab-field) were the only two important moderators.
- Job/organizational attributes and justice perceptions were weighed more heavily by real applicants.
- Recruitment predictors influence job attitudes and job acceptance intentions, which in turn influence job choice.
- Acceptance intentions are the best proxy for actual job choice but are imperfect.
- Actual job choice was studied infrequently and was poorly predicted.
- Job/organization attributes and fit influence intentions and behavior.
- Hard criteria are infrequently studied, and when they are, the relationships are much smaller.
- Demographics of both the applicant and recruiter seem to play a minor role.

Person-Environment Fit

Perceived person-environment fit (PE fit) is a central construct in recruitment.
The Chapman et al. (2005) meta-analysis, and another focused solely on fit (Kristof-Brown, Zimmerman, & Johnson, 2005), suggest multiple types of fit have broad implications for numerous criteria.
Research examines the meaning, measurement, and antecedents of subjective fit.
Kristof-Brown, Jansen, and Colbert (2002) found support for a three-level conceptualization of fit: person-job, person-group, and person-organization.
Cable and DeRue (2002) argued for three types of subjective fit perceptions:
- PO fit: Congruence between applicant's values and organization's culture.
- PJ fit: Congruence between applicant's competencies and job demands.
- NS fit: Congruence between applicant's needs and job rewards.
Discriminant validity was found for the three types of fit, and each type showed some unique relationships with different criteria.
PO fit related most strongly to organizational criteria, and NS fit related most strongly to job/career criteria.
PJ fit was unrelated to any of the criteria.
Cable and Edwards (2004) examined complementary fit (psychological need fulfillment) and supplementary fit (value congruence).
Both types of fit can focus on the same dimensions (e.g., participation, autonomy).
Complementary fit emphasizes amount, whereas supplementary fit emphasizes relative importance.
Both types of fit were related, each independently predicted the same criteria (and of approximately equal magnitude).
There are many different types of subjective fit perceptions, each providing unique prediction for specific criteria.
Understanding the consequences of fit perceptions will require correctly identifying the appropriate form of fit for the particular criterion.
Each type of fit may also have different antecedents.

Employer Brand Image

Employer image/reputation has important effects on recruitment outcomes (Saks, 2005).
Employer brand image is examined to encompass research on image, reputation, brand, and symbolic attributes (see Collins & Stevens, 2002).
Turban and Cable (2003) showed that an organization's reputation impacts applicant pool characteristics.
Firms with positive reputations increased number of applicants and influenced applicant behavior.
Both low- and high-ability applicants apply to organizations with favorable reputations, increasing recruiting costs.
More applicants allow for finer distinctions and more selective hiring of top talent.
Cable and Turban (2003) found applicants use firm reputation as signal about job attributes and source of pride.
Participants would accept a 7% smaller salary to join a firm with a highly favorable reputation.
Collins and Stevens (2002) borrowed from marketing literature to consider brand equity.
Brand image creates positive attitudinal reactions to organization and product's attributes.
Publicity, sponsorships, word-of-mouth, and advertising create positive brand image in early recruitment stages.
These practices (excluding sponsorship) influenced employer brand image, which in turn influenced applicant decisions.
Using multiple practices produced a stronger effect.
Collins and Han (2004) found recruiting practices and organizational information positively influenced applicant quality and quantity.
Advertising was the most important determinant of multiple measures of quality and quantity.
Lievens and Highhouse (2003) introduced the instrumental-symbolic framework to recruiting.
Instrumental attributes represent objective job and organizational attributes (e.g., pay, location).
Symbolic attributes represent subjective meanings and inferences (expressed in terms of trait or personality inferences).
Symbolic attributes provided incremental explanation of organizational attractiveness beyond instrumental attributes.
Symbolic attributes provided a more useful source of differentiation between competitors.
Slaughter, Zickar, Highhouse, and Mohr (2004) developed a measure of organizational personality and found individuals use organizational trait inferences to distinguish them from each other.
Organizational trait inferences were related to attraction.
Employer brand image offers sustained competitive advantage because it is rare, difficult to imitate, valuable, and cannot be substituted (Turban & Cable, 2003).
Fostering favorable employer brand image can be accomplished through advertising and similar practices (Collins & Stevens, 2002).
Favorable employer image influences both applicant and organizational-level recruiting outcomes (Collins & Han, 2004; Turban & Cable, 2003).
Employer brand image offers a way for organizations to differentiate themselves among applicants, even when they cannot compete in terms of location or wages.

Applicant Reactions

Applicant reactions research focuses on how applicants perceive and react to personnel selection practices (e.g., interviews, tests).
A review by Ryan and Ployhart (2000) identified two main streams of applicant reactions research:
- Perceptions and reactions that influence test-taker performance.
- Attitudinal perceptions and reactions to selection practices, with emphasis on fairness.
Test-taking motivation is a key construct, theoretically a proximal determinant of selection predictor performance.
Sanchez, Truxillo, and Bauer (2000) developed a construct-valid measure of test-taking motivation based on valence-instrumentality-expectancy (VIE) theory.
Research links demographic differences in selection perceptions and reactions to test performance.
Stereotype threat (negative stereotype about group's performance) has not been found to explain subgroup test performance differences in selection contexts.
A special issue of Human Performance in 2003 presented studies that manipulated stereotype threat for Blacks and women, calling into question the theory's applicability to employment testing contexts.
A high-stakes field study by M. J. Cullen, Hardison, and Sackett (2004) also found a lack of stereotype threat effect.
A special issue of the International Journal of Selection and Assessment in 2003 presented new theoretical developments in applicant reactions (e.g., fairness, decision making, expectancies, attribution process).
Bauer, Truxillo, Sanchez, Craig, Ferrara, and Campion (2001) developed a measure of procedural justice applicable to selection contexts.
A meta-analysis by Hausknecht, Day, and Thomas (2004) summarized 86 studies and examined applicant reaction predictors and outcomes.
Major findings from Hausknecht, Day and Thomas (2004):
- Selection procedures perceived as consistent, job-related/face valid, and explained to applicants will be perceived more favorably (enhancing motivation and performance).
- Job relatedness and face validity are most important perceptions.
- Demographics of the applicant show small relationships with the various outcomes.
- Personality-based perceptions are slightly stronger with some criteria (e.g., conscientiousness with test-taking motivation).
- Interviews and work samples are perceived most favorably, followed by cognitive ability tests, personality tests, biodata, and honesty tests. Resumes and references are perceived more favorably than cognitive ability.
- Nearly half of the studies were based on student samples, with others based on civil service positions.
There is good support linking applicant reaction predictors to intention and perceptual criteria, but links to objective criteria and actual decisions are much more limited.
It is disappointing that we still do not know whether applicant reactions influence applicant behavior and choices in the private sector.
It may be beneficial to study staffing managers’ reactions to staffing practices because these are likely to have a strong impact on their implementation decisions.

Internet Recruiting

Organizations use the Internet for recruiting (job search engines, organizational websites).
The Internet has become an important job search tool for applicants.
Cober, Brown, Keeping, and Levy (2004) presented a model describing how organizational Web sites influence applicant attraction.
The Web site's façade influences the affective reactions of job seeks, which in turn influence perceptions of Web site usability and search behavior.
Usability and search behavior influence attitudes toward the Web site, and search behavior and Web site attitude then influence image and familiarity.
These in turn influence applicant attraction to the organization.
Cober, Brown, and Levy (2004) identified common features and decomposed the Web sites into three dimensions: form (e.g., pictures, vividness, diversity images), content (e.g., culture, compensation information, fit messages), and function (e.g., interactivity, online applications).
The interaction between form, content, and function is critical.
The Internet can be an effective means to influence fit perceptions (Dineen, Ash, & Noe, 2002) and applicant attraction (Cober, Brown, Levy, Cober, & Keeping, 2003; Williamson, Lepak, & King, 2003).
Theoretical and classification work by Cober and colleagues (Cober, Brown, Keeping, et al., 2004; Cober, Brown, & Levy, 2004) is helpful in identifying the main features of Web sites that should be examined.
Little prescriptive advice (e.g., Web sites should be up-to-date, easy to navigate, aesthetically appealing, etc.).

Practical Recommendations and Implications for Organizational Effectiveness

Recruiting is most effective when an organization emphasizes fit information, provides details about the job and organization, selects and trains recruiters, treats applicants with fairness and respect, uses job-related procedures and explains the purpose of the selection process, articulates the right employer brand image, and ultimately creates a unified, consistent, and coherent recruiting campaign.
Organizations that use Web sites for recruitment should ensure they are aesthetically pleasing, easy to use, and provide the appropriate content for their purpose.
Saks argued,

Even though there has been a great deal of research on recruitment over the last thirty years (Breaugh & Starke, 2000), it is fair to say that a) there are few practical implications for recruiters and organizations, b) the practical implications that can be gleaned from recruitment research have been known for more than a decade, and c) the main practical implications are at best obvious and at worst trivial. (2005: 69)

A Table 1 lists a collection of key research-practice gaps in need of more attention.
There is a danger that recruitment research will fragment itself into very complex micro theories (e.g., complex theories specific to each kind and level of fit, employer brand image, Web site design, etc.), but the application will be lost in the details.
Despite impressive theoretical precision around some key aspects of recruitment research, there is still little information about such basic practical recruitment challenges as how to best recruit a diverse workforce/use targeted recruiting (Avery & McKay, 2006), how stage of recruitment affects practical effectiveness, how recruitment practices influence actual job choice, and perhaps most important, whether recruitment research offers a meaningful impact on organizational effectiveness.

Personnel Selection Best Practices

Personnel selection practices (e.g., interviews, ability and personality tests) continue to be the most active in staffing.
There are several comprehensive reviews of selection practices (e.g., Evers, Anderson, & Voskuijl, 2005; Schmitt, Cortina, Ingerick, & Wiechmann, 2003), as well as discussions of research and practical applications (Guion & Highhouse, 2006; Ployhart, Schneider, & Schmitt, 2006; Ryan & Tippins, 2004).
The review summarizes the major new developments.
Only those practices that have been the most active area of research or are likely to show the most important practical implications are discussed.

New Developments in Selection Practices

The main emphasis of recent cognitive ability selection research has been to identify ways of using cognitive ability that do not negatively affect racial diversity.
Strategies that may reduce subgroup differences include:
- Supplementing cognitive tests with noncognitive tests.
- Weighting criterion dimensions to de-emphasize task performance.
- Minimizing reading requirements.
- Enhancing face validity.
- Offering training and preparation opportunities.
New strategies are being evaluated, but employers must recognize that the sole use of cognitive ability may impair their ability to hire a diverse workforce.
Organizations will be tempted to avoid using the best selection methods if they negatively affect diversity.
Ryan, McFarland, Baron, and Page (1999) surveyed 959 organizations from 20 countries and found that compared to the grand mean of all countries, the United States was less likely to use cognitive ability tests and used fewer tests in general (see also Terpstra & Rozell, 1993).
For example, supplementing cognitive with relevant noncognitive predictors (or using situational judgment tests or assessment centers), ensuring the assessments have minimal reading requirements, and engendering favorable reactions might jointly reduce subgroup differences.

Personality

Research on personality in selection contexts continues to be active.
There have been advancements in understanding the mediators (Barrick, Stewart, & Piotrowski, 2002) and moderators (Barrick, Parks, & Mount, 2005) of personality-performance relationships.
Research has reconsidered traits less broad than those from the Five Factor Model (FFM).
A recent meta-analysis showed traits “narrower” than conscientiousness could provide incremental validity over global conscientiousness (Dudley, Orvis, Lebiecki, & Cortina, 2006).
There has also been research examining how personality contributes to team performance (e.g., Stewart, Fulmer, & Barrick, 2005).
Offsetting these advancements are nagging questions to some of the most basic issues.
For example, the now-classic meta-analysis by Barrick and Mount (1991) found personality traits (as measured on the FFM) demonstrated criterion-related validity for various criteria.
The argument comes down to whether the magnitudes of the validities are of practical benefit, or alternatively, why the validities are so “low.”
An entire special issue at Human Performance (2005) considered this topic.
Compared to incumbent settings, in applicant settings the mean scores are higher, the validity is about .07 lower, and the factor structures are highly similar (Hough, 1998; see Schneider & Smith, 2004).
Several approaches have been researched to help reduce faking, including the use of social-desirability scales, explicit warnings against faking, or use of response laten-cies to catch lying.
A notable measurement approach to reduce faking may come from research on conditional reasoning by James (1998), but this work has yet to see widespread application.
The Ryan et al. (1999) study described earlier found the United States is less likely to use personality tests relative to other countries.

Situational Judgment Tests (SJTs)

SJTs are predictor methods that present applicants with work-related situations.
Respondents are given several behavioral choices for addressing the situation and are then asked to indicate which options are most/least effective.
Research and practical applications of SJTs have exploded in the past several years.
SJTs show at least moderate validity ( $.26$ ; corrected $r = .34$ ; McDaniel, Morgeson, Finnegan, Campion, & Braverman, 2001), incremental validity over other predictors (Clevenger, Pereira, Wiechmann, Schmitt, & Harvey, 2001), and small to moderate gender/ethnic differences (Weekley, Ployhart, & Harold, 2004).
They appear applicable for selection at all levels, can be used to help prepare individuals for international assignments, and are useful for training and development.
Because the questions present realistic work situations, they tend to be received favorably by applicants and HR personnel.
An entire book on SJTs edited by Weekley and Ployhart (2006) had leading experts discuss these issues, and numerous theoretical and practical recommendations were offered.

Assessment Centers

Assessment centers present applicants with a variety of exercises (e.g., mock presentation, role-play) designed to measure multiple competencies.
They are predictor methods rather than assessments of homogeneous competencies.
Although they have long been known to demonstrate moderate to high levels of criterion-related validity, assessment centers have been plagued by an apparent lack of construct validity.
Specifically, scores on assessment centers demonstrate "exercise" factors rather than "construct" factors.
Recent research has made important strides in understanding why this occurs and showing assessment centers do, in fact, have construct validity.
Meta-analyses by Arthur, Woehr, and Maldegen (2000); Arthur, Day, McNelly, and Edens (2003); and Woehr and Arthur (2003) have summarized the validity, constructs, and exercises present in assessment centers.
The average assessment center uses approximately five exercises to measure 10 competencies, with the 6 most common being interpersonal skills/social sensitivity, communication, motivation, persuasion/influence, organization/planning, and problem solving.
Lievens (2002) has helped identify why assessment centers demonstrate exercise factors.
His research suggests that construct validity may be most determined by applicant behavior; applicants must demonstrate high consistency across exercises but also high variability across dimensions (see also Lance, Foster, Gentry, & Thoresen, 2004; Lance, Lambert, Gewin, Lievens, & Conway, 2004).
It is also true that convergent validity will be enhanced when trained assessors, particularly psychologists, conduct the evaluations.
Thus, even in the presence of exercise factors, assessment centers appear to have construct validity.

Work Samples

Work samples present applicants with a set of tasks or exercises that are nearly identical to those performed on the job.
It is believed that work samples provide one of the best ways to simultaneously achieve validity and diversity.
the Roth, Bobko, and McFarland (2005) found work samples show a corrected criterion-related validity of $.33$ , certainly good but much smaller than that often-cited $.54$ in the classic Hunter and Hunter (1984) publication (indeed, this puts them on par with SJTs).

Interviews

The interview continues to attract considerable research attention.
Posthuma, Morgeson, and Campion (2002) published an extensive narrative review that does an excellent job of organizing the massive interviewing literature.
Research has clearly found that structured interviews are more predictive than unstructured interviews or even interviews with less structure (Cortina, Goldstein, Payne, Davison, & Gilliland, 2000).
A meta-analysis by Huffcutt, Conway, Roth, and Stone (2001) identified seven latent dimensions most assessed by interviews: cognitive ability, knowledge and skills, personality, social skills (e.g., leadership), interests and preferences, fit, and physical abilities and attributes.
Panel interviews composed of diverse interviewers may help further reduce subgroup differences, although the findings are quite complex (McFarland, Ryan, Sacco, & Kriska, 2004).
Schmidt and Zimmerman (2004) used metaanalysis to show that it takes about four unstructured interviews to equal the reliability of one structured interview.

Summary

Research on these and other selection practices will continue.
One must wonder if part of the appeal in selection methods (SJTs, assessment centers) is their generally lower racial/gender subgroup differences and greater face validity.
Research may be close to solving the construct validity question for assessment centers, but it will surely dominate the study of SJTs for the next several years (Schmitt & Chan, 2006).
A comment on this entire line of research is that researchers tend to limit their focus to correlations with various individual-level criteria, convergent/discriminant validity, and subgroup differences.

Selection Using the Internet

Nearly every major staffing firm has adapted some form of Internet-based testing, and many organizations have already migrated from paper to Web-based selection.
The rush to use this delivery platform is appealing: efficiency and cost savings, ability to administer the test globally in real time, and standardized scoring and administration.
The issues involving Web-based selection are quite different from Web-based recruitment, in large part because there is more legal scrutiny with selection practices.
Moving from a paper format to a Web format requires one to demonstrate the equivalence of the two formats (e.g., Potosky & Bobko, 2004).
If using Web-based testing, one must then choose between proctored versus unproctored Internet testing, and there is no clear professional consensus over whether unproctored testing is appropriate or feasible (see Tippins et al., 2006, for an sampling of opinions).
There are a variety of legal issues surrounding Internet testing, but many of the same basic issues present with traditional employment testing are present with Internet testing (Naglieri et al., 2004).

Practical Recommendations and Implications for Organizational Effectiveness

Organizations wishing to best balance diversity and prediction should use selection methods such as assessment centers, work samples, or SJTs.
One approach that also enhances validity is to include a battery of cognitive ability and personality predictors (as necessitated by job demands).
There appear to be meaningful benefits to administering assessments over the Internet, but we know very little about the consequences of using Web-based assessment.
The safest thing an organization can do (for now) is to collect the content or criterion-related validity evidence necessary to support the Web-based procedure, until a research database is developed that can inform such practices.

Multi-level Staffing: Linking Individual Staffing to Organizational Effectiveness

The reviews of recruitment and selection practices both identified a need for research showing business unit value/organizational impact.
This is interesting given the most basic staffing assumption, one described in nearly every textbook written on the subject, is that recruiting and hiring better employees contributes to organizational effectiveness.
However, there is actually little direct, empirical evidence testing this assumption (e.g., Ployhart, 2004; Saks, 2005; Taylor & Collins, 2000).
Utility analysis may be helpful to estimate these effects, but they are only estimates that are limited to monetary outcomes and are frequently discounted by managers (Schneider, Smith, & Sipe, 2000).
Practitioners and HR managers often have to go well beyond validity (and even utility/monetary estimates) to make a case that staffing adds strategic value to the firm.
Micro (individual)-level research examines how individual differences (knowledge, skills, abilities, and other characteristics; KSAOs) contribute to individual performance but assumes (or only estimates how) individual differences contribute to organizational value.
Macro (organizational or business unit)-level research examines how HR practices (e.g., staffing) contribute to organizational performance but assumes that these practices have an effect because of their influence on employee KSAOs.
This is known as a cross-level fallacy in multi-level research and occurs when researchers inappropriately generalize their within-level findings to higher or lower levels of analysis (Rousseau, 1985).
Staffing may be one of the last holdouts to develop such multi-level theory.
Schneider et al. (2000) strongly conveyed a need for multi-level staffing research, suggesting the very relevance of staffing may be ignored because of an inability to show unit-level value.
Therefore, the next section introduces basic multi-level concepts critical to multi-level staffing, followed by multi-level staffing models.

Multi-level Theory

Organizations are inherently nested and hierarchical, for example, individuals are nested within business units such as departments or stores, which are in turn nested within the firm.
Multi-level theory argues that ignoring such hierarchical structures can cause misleading interpretations and generalizations of within-level research findings (with cross-level fallacies being just one example).
One important implication is that observations (e.g., employees) within a unit (e.g., store, organization) are likely to share similarities on particular KSAOs.
To connect levels, multi-level theory describes theoretical processes for both contextual effects and emergent effects.
- Contextual effects are “top-down” effects from higher to lower levels (e.g., changing an organization’s HR practices changes the behavior of individual employees).
- Emergent effects are “bottom-up” effects from lower to higher levels.
Composition models of emergence theorize that there is such high similarity (homogeneity) among lower level observations (employees) that the within-unit scores create a distinct aggregate-level construct.
Compilation models of emergence theorize that variability (heterogeneity) among lower level observations (employees) represents a unique higher level construct.

Multi-Level Staffing Models

Multi-level staffing models are based on the integration of traditional micro-level staffing research with macro-level strategy and SHRM research.
Multi-level theory is used to fuse these disciplines and explicate how individual differences contribute to the formation of unit differences.
Human capital emergence represents the multi-level processes through which individual-level KSAOs become organizational or business unit−level human capital.
Organizational-level human capital contributes to the organization’s performance, such that firms with higher quality human capital will outperform those with lesser quality human capital.
This is known as human capital advantage in the macro literature (e.g., Boxall, 1996).
These points represent some important areas of departure between multi-level staffing models and traditional staffing models.

Empirical Support

Empirical support for aggregate-level human capital as a means to differentiate units has been found in several studies (Jordan, Herriot, & Chalmers, 1991; Schaubroeck, Ganster, & Jones, 1998; Schneider, Smith, Taylor, & Fleenor, 1998).
Support for a multi- level model of human capital emergence was provided by Ployhart, Weekley, and Baughman (2006), who found human capital emergence (operationalized via personality) was hierarchical such that emergence was stronger at lower than higher levels.
Four types of employment, with corresponding HR configurations, were identified: knowledge based (commitment-based HR practices), job based (productivity-based HR practices), contract (compliance-based HR practices), and partnerships/alliances (collaborative-based HR practices).

Practical Recommendations and Implications for Organizational Effectiveness

Multi-level staffing models do not negate the importance of single-level recruitment and selection research.
Rather, they seek to extend this work by articulating the linkages between individual differences and organizational/business unit differences.
Multi-level staffing also offers the opportunity to advance staffing theory.

Neglected Questions That Shouldn’t Be

First, why do managers so often fail to believe in our technology and science?
Second, how does staffing contribute to reinforcing/changing/articulating organizational culture, climate, values, personality, and vision?
Third, what are the consequences of outsourcing staffing?
Fourth, we know very little about the implementation, use, and effectiveness of staffing practices across (not within) cultures (Ryan et al., 1999, is a notable exception).
Finally, are findings based on civil service organizations generalizable to private sector organizations?

Staffing at the Dawn of the 21st Century

Staffing sits in a curious position at the dawn of the 21st century: Economic, societal, and cultural changes make organizational success and survival dependent on staffing, but many organizational decision makers and even organizational scholars fail to recognize staffing’s value.
Staffing should reign supremely strategic in the war for talent and sustained competitive advantage, but it is incumbent on staffing researchers and practitioners to show the organizational value of their science and practice (a concern of HR more generally).
Research on traditional recruitment and selection practices is important and should continue, but this by itself seems unlikely to increase strategic value.
Multi-level staffing research and models were offered as one mechanism for conveying business unit value.
Every single organization in the world uses some form of staffing procedure, but there is no guarantee they use them optimally or even appropriately.