Estimating the Value
of the Indirect
Benefits of New
HR technology
By Kevin D. Carlson, Ph.D., Pamplin College of Business,Virginia Tech
W
hether organizations continue to
invest in emerging HR technologies depends to an increasing degree on
our capacity to develop reasonable estimates of the value offered by new functionality. Organizations’ purchases of HR
technology have generally been aimed at
cost reduction. However, the impact of
new investments is likely to shift from
cost reduction to organization enhancement and current cost-benefit analysis
approaches must be modified to meet
the needs of these types of analyses.
This article examines one of the more
difficult tasks in HR technology costbenefit analysis — estimating the value
of the indirect benefits of new functionality. These analyses are more challenging because such enhancements typically do not impact costs or revenues
directly. Instead their impact occurs indirectly through their effects on important intermediate outcomes, i.e., better
decision making, or is contingent upon
other factors that may or may not be under organizational control, i.e., the economy. The difficulty of converting indirect
benefit estimates to a dollar metric has
limited their role in HR technology investment decisions. It is not uncommon
for “soft savings” to be relegated to the
narrative supporting an investment
analysis that is otherwise based solely
on estimates of direct cost reductions.
For good reason, many managers consider these savings cautiously due to
their indirect relationship to revenues
and costs. That does not mean, though,
that these benefits are any less important to the organization. In fact, ignoring
22
July/August 2004 • IHRIM Journal
them in HR technology investment
analyses could result in decisions not to
approve valuable investments. As a result, we need to adapt the techniques
used to analyze HR technology investments to meet these new requirements.
The challenge is to provide an estimate
of the value of soft or indirect benefits
and to do so in a way that can be justified to skeptical decision makers.
This discussion of tools and techniques assumes that the benefits resulting from new HR technology have been
identified and that some of those benefits have indirect or contingent effects.
Methods for identifying sources of value
in HR technology/HRIS investments are
discussed in an earlier IHRIM Journal article (Carlson, 2004). Once a list of indirect
benefits has been generated, estimating
the value of those benefits can begin. The
approach described here involves three
steps: estimating indirect benefit magnitude, mapping indirect benefits to cost or
revenue changes, and valuing indirect
benefits. Each of these processes will be
addressed in order. I will close with some
comments on packaging indirect benefits
for decision makers.
ESTIMATING INDIRECT
BENEFIT MAGNITUDE
It is relatively easy to estimate the
dollar-valued impact of gaining a new
sales contract or canceling a software
support contract. For indirect benefits,
though, constructing dollar-valued estimates is more difficult. It is possible to
estimate the dollar value of indirect benefits directly, but in most instances,
breaking this task into the three steps
mentioned above — estimating benefit
magnitude, mapping benefits to cost or
revenue changes, and then converting
magnitude estimates to dollar values —
simplifies estimation and justification.
By separating these steps, we can begin
to understand more of the factors that
influence the value of indirect benefits,
but, and perhaps more importantly,
when they are likely to occur. Also, since
magnitude and value are often driven by
different factors, separating these decisions provides a better framework for
post-implementation evaluations. Both
benefit and value estimates are open to
objective review.
The objective of HR technology cost
benefit analysis is to develop the best
possible estimate of the likely effect of
the new functionality. Using the metric
that is most familiar or most comfortable
to those developing the estimate is likely
to improve accuracy. For instance, if new
functionality is predicted to reduce
turnover, being able to first estimate the
magnitude of the expected change using
“turnover rates” and then determine the
value of those differences is likely to produce better estimates than if we required
decision-makers to estimate the dollar
impact of the expected effect in a single
step. Analysts are, therefore, free to
choose the unit of analysis they feel is
most appropriate, whether that unit is
people hours, new hire potential, rework
and scrap units reduced, or percent increase in market share. The objective is to
choose the metric that will result in the
most accurate estimate of the size of the
FEATURE
benefit possible. Once that metric has
been chosen, three approaches for estimating benefit magnitude exist: direct estimation, benchmarking and internal assessment. Which method is most
appropriate depends on the amount of
specific information that is available to
the organization.
Direct estimation is the simplest of
all three methods. It is quick and easy to
perform. It relies solely on the expertise
of the analyst or a subject matter expert
to “estimate” the expected magnitude of
the benefit. Direct estimation is most
appropriate when project scopes are
small, compliance or risk mitigation is a
primary investment justification, other
substantial sources of direct cost reduction or direct revenue enhancement exist, or no other method for estimating
benefit magnitude is available. The primary limitation of direct estimation is
that the accuracy of the analysis depends on the expertise of the estimator.
When several equally qualified subject matter experts are available, collecting independent estimates from each
expert and using the average of these estimates is recommended. In addition, it
can be useful to require that experts articulate the rationale for their estimates.
This not only assures that experts are
thoughtful in the preparation of their estimates, but an analysis of the assumptions or expectations contained in these
rationales can also be used to help improve the accuracy of future estimates.
Benchmarking involves using data
from other firms as the basis for estimates. Data about the magnitude of benefits achieved in other firms can be used
to estimate the magnitude of benefits a
firm could expect if they implement similar functionality. The advantage of
benchmarking is that it allows an organization to build on the experiences of others. This data provides evidence that a
specific outcome can occur, as well as its
potential magnitude. Howes (2002) offers an insightful example of how benchmarking data can be used to estimate
how much reduction in turnover an organization might expect. In this example,
benchmark data about industry-wide levels of turnover is used to construct estimates of the potential for improvement
in turnover that might be possible for a
given organization — if an organization
has high turnover relative to industry
standards, it has the potential for greater
improvement than might be expected for
other firms in that industry.
Benchmarking information of various
types is becoming more widely available
from a number of sources, e.g., Cedar
Group, Gartner Group, Hackett Benchmarking, Saratoga Institute and others.
In addition, organizations can conduct
their own benchmarking studies to
gather specific data from targeted firms
that may not be readily available from
third-party sources. Benchmarking is
preferred over direct estimation for
larger, more costly projects where invest-
ment risks are greater. Benchmarking is
also useful when organizations have limited experience with the targeted functionality or when there is no access to local data. The primary disadvantage of
benchmark data is that the experiences
of other organizations may not generalize to your firm or business unit.
Internal assessment involves the use
of a firm’s own internal metrics or other
forms of firm-specific data as the basis
for estimates. Internal assessment is
best when investment scopes are large
and direct estimation and/or benchmarking suggests benefits may not be
dramatically higher or lower than costs,
Table 1. Different Approaches to Estimating Benefit Magnitude.
IHRIM Journal • July/August 2004
23
FEATURE
Figure 1. Conceptualizing the Effects of Turnover.
e.g., less than +/- 30 percent. Internal assessment requires that the organization
possess the capabilities to gather the
data about their own processes necessary to support these analyses. An advantage of integrated information systems in organizations is that the
marginal costs of assessments are
greatly reduced, permitting cost effective assessments of a wide range of organizational outcomes. Internal assessments offer the most precise estimates
of the costs and performance of existing
or newly implemented processes. Internal assessments, though, may only be
able to provide a portion of needed data.
That is, an organization may be able to
gather accurate data about current
process outcomes, but may need to rely
on benchmark data or direct estimates
of the new outcomes expected to result
from the new functionality.
Even though possessing integrated
information systems can reduce the
marginal costs of assessments, conducting internal assessments and evaluating
the data they produce are not costless
activities. Internal assessments generally result in higher costs than direct estimation and, depending on the nature
of the assessment, could result in higher
decision support costs than benchmarking. And as noted above, internal assessment is only possible when the or-
24
July/August 2004 • IHRIM Journal
ganization has experience with a given
form of functionality. It is not possible to
assess the effects of new functionality
that have not previously been implemented anywhere in an organization.
These three approaches are recapped
in Table 1. The ideal approach for estimating the magnitudes of indirect benefits in most analyses is to use a combination of these three approaches. This
permits each benefit to be estimated using the method that is most appropriate
given the availability of data and the investment’s cost, risk and opportunity
characteristics. For high stakes investment decisions, using multiple methods
to develop estimates can provide additional insight and increased confidence.
MAPPING INDIRECT BENEFITS TO
REVENUES AND COSTS
In some instances, the metric of
choice may not be one that is easily or
unambiguously tied to reductions in
costs or increases in revenues. Estimating the value of indirect benefits requires that the analyst first be able to articulate how the indirect benefit is
linked to an actual reduction in costs or
an increase in revenues. For example,
let’s assume that an indirect benefit of a
proposed investment is reduced
turnover and in the process of estimating benefit magnitude just described we
predicted that implementing new HRIS
functionality will result in a reduction in
voluntary turnover from 10 to five percent for a targeted group of jobs. Since
the effect of turnover on costs and benefits is indirect, we need to understand
how reducing turnover is expected to influence organization revenues or costs
in order to translate our five percent reduction in turnover to other metrics that
are more closely associated with
changes in costs and revenues.
Employee turnover is a good example because it impacts costs and revenues in several ways, some of which are
depicted graphically in Figure 1. The departure of an employee can increase
costs because it may require the organization to engage in a new recruitment
and hiring cycle and the new employee
is likely to require training. But there are
other effects as well. For instance, the
loss of an employee in a position critical
to the day-to-day functioning of the organization requires that efforts be made
to cover the work responsibilities in that
employee’s absence. How an organization chooses to cover those responsibilities influences the magnitude of the net
loss of contribution that results from the
vacancy. There will be salary savings for
the vacant position, but the cost of temporaries, shifting other employees off
their primary assignments, and the op-
FEATURE
portunity costs that result from using
less than fully effective temporary or
over-extended employees also must be
considered. The total loss of this contribution is represented in Figure 1 by the
region A*B — the value of the daily loss
of contribution multiplied by the number of days the position remains vacant.
Also, as noted in Figure 1, turnover impacts the contribution the organization
derives from a position. Perhaps the departing employee was a poor performer
and replacing them will actually result in
a net gain in average long-term effectiveness for that position, i.e., C in Figure 1.
Even with training, new hires will most
likely require some time on the job before they will become fully effective;
their effectiveness will increase as they
gain expertise. But during this time, contribution will be less than would have
been experienced had there been no
turnover in the position, i.e., E in Figure
1. Each of these intermediate outcomes
can be tied to a specific cost or revenue
effect through one or more links.
In this example, a comprehensive estimate of the impact of the expected reduction in turnover is represented by the
sum of the estimates of each of these
components. In some cases, and for specific types of benefits, these relationships may seem quite complex. Do not
be discouraged or dissuaded from being
thorough. Understanding exactly how
these changes are projected to impact
the organization may yield important
new insights about these intermediate
outcomes and contingent factors that
put managers in a position to ensure the
success of HRIS investments. Understanding which factors are affected by
the investment can also aid in further refining magnitude estimates and are essential to estimating value.
ESTIMATING THE VALUE OF
INDIRECT BENEFITS
Direct revenue enhancements or direct cost reductions are typically estimated in dollars, so their value is provided by the magnitude estimate.
Estimates that are developed in other
metrics must be converted to dollars. For
indirect benefits that can be tied more
directly to cost reductions or revenue enhancements, i.e., new products developed or market share increased, the task
is somewhat more difficult, but can be
done. It requires estimating the strength
of the relationship between the change
in intermediate outcomes and changes
in revenue or cost, i.e., each new account
will generate US$50,000 in gross profit
annually; reducing scrap by five percent
will save US$10,000 per month. For other
outcomes, like employee time saved,
these conversions are somewhat more
difficult to conduct. I’ll focus on this last
category and briefly discuss two methods for estimating the value of employee
time that have slightly different purposes
— average employee contribution and
utility analysis.
These methods are alternatives to a
third practice that is not recommended
— estimating an employee’s value as
equal to their cost to the organization. In
nearly all cases, employee cost dramatically underestimates the average employee’s contribution. We know this is
true because most organizations are
profitable. For that to occur, each individual in an organization, on average, contributes enough value to return the cost
of their wages, benefits, and the equipment and facilities they employ in their
job (more on this in a moment) in addition to covering taxes and accounting for
profit. This helps explain why downsizing
does not always improve financial results. Downsizing only makes sense
when the contribution of the employees
eliminated is less than their cost in
salary and benefits to the organization.
In the turnover example developed
earlier, estimating the value lost while a
position remains open (represented by
the area A * B in Figure 1) requires an estimate of the value of the average daily
contribution made by an employee. The
average contribution approach argues
that the contribution of the average employee (AEC) in an organization is equal
to total gross profit divided by the number of employees (full-time equivalents).
Average employee contribution is not a
metric that most organizations track,
though a measure of average daily gross
margin, i.e., revenue minus cost of goods
sold, generated per sales person would
be an example of this type of measure.
AEC = (Net Revenues – Cost of Goods
Sold) / Number of FTE employees
By definition, in profitable organizations this number will be substantially
above labor costs. Dividing this number
by the number of workdays in a year, i.e.,
252, produces an estimate of average
daily contribution. As noted in the equation above, this method does not reduce
contribution attributable to employees
by the amount of capital expenditures
and other non-employee related expenses — the equipment and tools that
aid employees in doing their jobs. The
reason is these tools are used to enhance employee contribution. The tools
enhance what employees can do, but
they do not generate contribution on
their own. That is, if you take away the
employees that use the tools and support the equipment, the contribution
goes with them. The tools provide no independent and unique contribution to
the organization absent the employee.
Average employee contribution can
be used to estimate the average annual
contribution of an organization’s employee. Obviously, average contribution
can and does differ across jobs. Thus, organizations may want to adjust this
number up or down for specific jobs to
recognize differences in contribution potential. Jobs that more closely support
the organization’s mission, offer jobholders broader authority to influence
the work of others, and offer greater autonomy for choosing how and when
work will be accomplished are likely to
offer above average levels of contribution. The advantage of this method is
that it establishes a baseline contribution value consistent with the actual financial performance of the organization.
Average contribution estimates,
though, provide little guidance in estimating differences in contribution between employees holding the same or
similar jobs. This is represented in the
turnover example in Figure 1 by the difference between the values “C” and “D”
— the differences in contribution for two
different employees who might hold this
position. Differences in contribution can
be developed using internal assessment
by directly examining variance in the
outcomes produced by a large number
of individuals holding equivalent positions. This can be accomplished most
readily for jobs where individual production rates can be monitored, e.g., sales,
IHRIM Journal • July/August 2004
25
FEATURE
transaction processing, and some manufacturing settings.
However, not all settings allow for the
development of these data. There may
not be a large number of people doing
similar work or a job’s content may
change frequently. In these instances,
utility analysis techniques can be useful.
There are several versions of utility
analysis; I prefer methods based on the
“Brodgen-Cronbach-Gleser” model described by Boudreau (1990). Utility
analysis is a set of decision-making
tools and techniques used to estimate
the value of human resource management programs. It is based on the assumption that employees are not perfectly interchangeable; that is, selection,
placement and training are important
because there are differences in the contribution possible from a position that
depends on the characteristics of the
employee holding that position. Utility
analysis is also useful for a second reason: it provides a mechanism for converting differences in estimated performance to dollars. In utility analysis, the
value of differences in job performance,
i.e., contribution, among individuals is
captured in a metric called the standard
deviation of performance in dollars
(SDy). Standard deviation of performance represents the value of one standard deviation difference in average job
performance. This is roughly equivalent
to the expected difference in annual
contribution made by an employee
whose performance would be deemed to
be at the 85th percentile of all people
who might hold that job as compared to
an employee at the 50th percentile. If you
can estimate differences in job performance, or performance potential in the
case of selection, you can combine these
data with an estimate of SDy for that position to estimate differences in contribution between employees.
The most common method for developing SDy estimates has been to ask
several supervisors to produce direct estimates of these values and then average
the individual estimates. An alternative
method of estimating SDy has evolved
from reviews of such estimates and instead estimates SDy based on the pay
level assigned to the job. To some extent, differences in contribution are a
function of the level of autonomy and
26
July/August 2004 • IHRIM Journal
authority individuals are given. Individuals with higher autonomy and authority
have greater freedom to determine what
outcomes they will pursue and the
methods they will use to achieve them.
This increases the potential variability in
outcomes; if they choose well they can
greatly increase outcomes, if they
choose poorly, outcome levels can drop
dramatically as compared to jobs where
individuals have little discretion and
must follow prescribed procedures. Not
surprisingly, most job evaluation methods also recognize these factors and,
consequently, jobs with these characteristics are assigned higher salaries. It appears that, in the absence of more specific information, a baseline estimate of
SDy can be developed by multiplying annual salary times .40 (Judiesch, Schmidt,
& Mount, 1992; Schmidt & Hunter, 1983;
Schmidt, Hunter, McKenzie, & Muldrow,
1979). So, for example, a position with
an average salary of US$30,000 per year
might be expected to produce an SDy of
US$12,000. Thus employees with job
performance at the 85th percentile for
that job would be expected to contribute
US$12,000 more per year than employees at the 50th percentile.
The value of improved job performance could then be estimated by determining the differences in expected
contribution levels. This can be calculated by estimating an individual’s performance score using performance evaluation data or performance predictor
scores in selection contexts and dividing
it by the standard deviation of all scores
on that measure. This produces a standardized score (Zx). Differences in estimated contribution between individuals
can then be determined by finding the
difference in their standardized performance scores and multiplying it by the
estimated value of SDy for that position.
Difference in Daily Contribution =
[(Zx2 – Zx1) * SDy] / 252
In our example, the 85th percentile is
equivalent to a standardized score of 1.0
and the 50th percentile equates to a
standardized score of 0.0. The calculation or difference in daily contribution
then is as follows:
Difference in Daily Contribution = [(1.0
– 0.0) * US$12,000] / 252 = US$47.62
This value represents a slight overstatement of the actual value because
nearly all methods of estimating employee performance score have some
measurement error. A more accurate
value can be obtained by reducing this
estimate by the estimated validity of the
performance evaluation, i.e., the correlation between rated performance and actual performance. For selection contexts, the reduction may be more
substantial. Current research suggests
the validity of selection data for predicting performance ratings may be no
higher than .5 to .7 (Schmidt & Hunter,
1998; Hough & Oswald, 2000). Because it
uses standardized scores, utility analysis
techniques are useful for comparing the
contributions of two individuals (or
groups of individuals), but, unlike AEC,
it cannot be used to develop estimates
of an individual’s absolute contribution.
Obviously, the two methods described here offer relatively coarse tools
for estimating the value of individual
contributions in organizations. However,
both are theoretically sound, and perhaps more importantly, they are transparent. Analysts and decision-makers
are likely to disagree on the inputs entered into utility analysis or AEC calculations. If so, they can easily be changed
and the effects of different values examined. But the critical point is that these
values can and should be estimated.
Over time and with experience our
methods will improve. But not using
these or other methods to estimate the
value of changes in employee contributions is not appropriate. These contributions do exist and are a critical component of HRIS investment analyses.
VALUING TIME SAVED
Many types of HR technology can
save employee time, making it a common component of HR technology costbenefit analyses. When new functionality will save enough time to make it
feasible to reduce the number of employees or reduce overtime expenses,
time saved is a source of direct cost reduction. However, more often HR technology saves time in smaller increments
that do not permit direct savings. That
is, the amount of time saved does not
FEATURE
permit whole positions to be eliminated. In these circumstances, timesaved is an indirect benefit. The value of
the time saved actually depends on what
value generating activities those individuals conduct during the time made
available to them. For example, if the
implementation of self-service functionality reduces workload, but does not
lead to headcount reductions, the new
functionality might have tremendous organization value if those saved hours are
used to improve the effectiveness of recruitment efforts or some other value
generating activity.
Time saved, though, may not always
have value. Consider a situation in
which an individual engages in an activity that requires five minutes every day,
but the application of new HRIS functionality is estimated to cut this time
from five minutes to one minute. What is
the value of the four minutes saved each
day? Generally, larger blocks of time are
more easily employed in value enhancing activities. Consider your own use of
time during the day. Could you constructively employ an additional minute of
time each day? In most cases, we already
have several of these minutes in our
schedule that, because of the ebb and
flow of daily events, are difficult to use
productively. Therefore, it is questionable whether most employees can consistently use short periods of time, i.e.,
blocks of less than five minutes,, productively. Thus it may be very difficult to
generate value for HR technology that is
expected to save time, but does so in
many small increments.
PACKAGING THE ANALYSIS
FOR DECISION-MAKERS
Once you have estimated the magnitude and value of all sources of benefits,
the focus shifts to packaging the analysis
for consideration by decision-makers.
This includes deciding what data to include and how that data should be organized. A table outlining the value and
timing of costs and revenues is likely to
be the central focus of the analysis. Some
experts encourage limiting the number of
sources of benefits presented to decision-makers in order to simplify the presentation and the required justifications.
While being able to make your case on a
single page is beneficial, there are several
advantages to including all cost and benefit components that “materially” influence the likely outcomes of the investment. First, this offers the most
complete, best estimate of the value of
the investment, thereby giving decisionmakers the best chance to make an appropriate investment decision. Second, it
provides the decision-maker with a fuller
understanding of the investment and the
impact of the investment on the organization. Particularly with respect to indirect benefits, contingent actions taken by
managers are likely to influence the extent to which the estimated benefits will
be achieved. Making decision-makers
aware of these contingencies can help
enlist their assistance in assuring each
investment’s future success.
report averages and do not report variance data. In some instances it may be
possible to request the standard deviation associated with each benchmark
value from the reporting organization.
With internal analysis, variance estimates can be calculated from the
archival records of existing processes before and after implementation. In all instances, an estimate of the standard deviation of outcomes could then be used
to provide a range of the most likely outcomes. However, remember that in the
absence of compelling evidence to the
contrary it is the mean value that represents your best estimate of expected outcomes and it is the mean value that
should be the focus of your analysis.
CONSERVATISM
THE ROLE OF VARIANCE
IN ESTIMATES
Since the estimates produced for
cost-benefit analyses are necessarily forward looking and may depend on events
outside of management control, actual
outcomes are likely to deviate from those
estimated. For example, let’s say that on
average you expect a reduction of two
hours processing time per transaction,
but some transactions will take longer,
some shorter. The actual number will
vary depending on the mix of transaction
types, operator expertise, and other job
requirements. The primary estimate of
interest is the overall average expectation. However, particularly for indirect
benefits, it is useful to develop expectations about the range and distribution of
possible outcomes. Estimating lower
and upper bound estimates or standard
deviations for magnitude and value estimates is useful auxiliary information that
can help convey expectations about potential variability in outcomes. Variance
estimates can be developed for each of
the three estimation methods. For direct
estimation, local experts could produce
estimates of the range and likelihood of
various outcome levels. In addition, multiple estimators could be used if two or
more equally knowledgeable individuals
exist. Each could be instructed to estimate a target value and upper and lower
bounds that could then be averaged to
develop overall estimates. Variance estimates for benchmark data are more difficult to acquire since most sources only
Analysts are sometimes encouraged
to develop conservative estimates. This
can be accomplished by scaling back,
i.e., reducing, estimates of magnitude
and/or value or not including certain
benefits when packaging the analysis for
decision-makers. Being conservative is
appropriate for decision-making regarding the investment decision, not the
analysis. By developing the best estimate possible and then supplementing
that estimate with projected upper and
lower bound estimates you provide decision-makers the opportunity to determine what they believe is the most appropriate level of conservatism. It is
important to recognize that even if you
have already used conservative values in
your estimate, decision-makers are
likely to add their own additional conservatism in their decision-making. If
both the analyst and decision-maker are
conservative, it is likely that decisions
may be based on estimates that unnecessarily understate the actual benefits
that can result from the investment.
CONCLUSION
Accurately portraying the costs and
benefits of new HR technology will play
an important role in organizations’ HR
technology investment decisions for the
foreseeable future. This is reflected in
the collective emphasis of vendors on
payback rates, return on investment, or
reduced cost of ownership in their promotional materials. Renewed interest in
detailed investment analysis is healthy
IHRIM Journal • July/August 2004
27
FEATURE
and should be embraced by HRIS managers. In addition to the primary intended outcome of making better purchasing decisions, detailed cost and
benefit analyses of HR technology investments are also likely to result in a
more complete understanding of the implications of each investment decision.
Thorough analyses are likely to surface,
implementation contingencies that,
once known, can be incorporated in order to increase the chances that HR
technology investments actually do improve organization effectiveness.
REFERENCES
Boudreau, J. “Utility Analysis for Decisions in Human Resource Management.” In M. D. Dunnette & L. M. Hough
(Eds.), Handbook of Industrial and Organizational Psychology (Vol. 2, pp. 621-752). Palo
Alto, CA: Consulting Psychologists
Press. 1991.
Carlson, K. D. “Justifying HRIS Investments Post-Y2K: Identifying Sources of
Value.” IHRIM Journal, 3(1), 21-27. 2004.
Hough, L. M., & Oswald, F. L. “Personnel Selection: Looking Toward the Future—Remembering the Past.” Annual
Review of Psychology, 5, 631-664. 2000.
28
July/August 2004 • IHRIM Journal
Howes, P. “Calculating the ROI for an
HRIS Business Plan.” IHRIM.link, 12-15.
2002, February/March.
Judiesch, M. K., Schmidt, F. L., &
Mount, M. K. (1992). “Estimates of the
Dollar Value of Employee Output in Utility Analyses: An Empirical Test of Two
Theories.” Journal of Applied Psychology, 77,
234-250. 1992.
Schmidt, F.L., & Hunter, J.E. “The Validity and Utility of Selection Methods in
Personnel Psychology: Practical and
Theoretical Implications of 85 Years of
Research Findings.” Psychological Bulletin,
124, 262-274. 1998.
Schmidt, F. L., & Hunter, J. E. “Individual Differences in Productivity: An Empirical Test of Estimates Derived from Studies of Selection Procedure Utility.” Journal
of Applied Psychology, 68, 407-414. 1983.
Schmidt, F.L., Hunter, J.E., McKenzie,
R.C., & Muldrow, T.W. “Impact of Valid
Selection Procedures on Work-force Productivity.” Journal of Applied Psychology, 64,
609-626. 1979.
Kevin D. Carlson is an associate professor of
management in the Pamplin College of Business
at Virginia Tech where he teaches courses in human resources and productivity management
and a course entitled “Applied Human Resource
Information Systems” that focuses on the effective utilization of HRIS. Dr. Carlson’s work has
been published in the Journal of Applied Psychology, Personnel Psychology, Journal of Management, Personnel Review and IHRIM Journal. He has presented papers at the meetings of
the Academy of Management, the Society for Industrial and Organizational Psychology and the
National Institute for Staff and Organizational
Development. He is a member of the Carolinas
Chapter of IHRIM and has presented professional development seminars on HRIS cost-benefit analysis at the IHRIM Annual Conference.
His current research interests include modeling
the determinants of performance, knowledge
structures and the development of competence,
self-regulatory activities, recruiting and selection, and human resource information systems.
His mailing address is 2094 Pamplin Hall,
Blacksburg, VA 24061. He can be reached at
[email protected].