Research Note: Trade Name Franchise Membership as a Human Resource Management Strategy: Does...

Human resource management (HRM) is an important source of competitive advantage. However, recent reports document that HRM, and in particular employee training, is less likely to occur in small firms. One possible remedy of special interest to small retailers is trade name franchise membership.

Trade name franchises, or buying groups, claim to offer small firms advantages not otherwise realizable, including bulk buying discounts, brand name recognition and workforce training. We report results from our study of over 300 small and medium-sized hardware stores, which tested the effects of employee training and trade name franchise membership, on small firm performance. Trade name franchisees achieve higher mean performance; however, the contribution to performance made by franchisees' training does not appear to be significantly better than that of independent stores.

The retail landscape has been forever changed. A whole new species, the big box retailer, has entered and increasingly dominates the retail arena. In sector after sector, the big box applies its "category killing" logic leaving many retail observers to wonder whether the days of the small 'Mom and Pop' retail operation are numbered (Ehrenfeld, 1995). Not all observers are equally gloomy about the small firm's prospects, however. Proponents of collective alliances known as trade name franchises, or buying groups (Hardy & Magrath, 1987), assert that through inter-organizational alliances small firms can realize advantages not otherwise accessible. By joining with other similarly constrained operations, the argument goes, small firms can realize economies of scale in purchasing, gain benefits of franchise brand name recognition, and receive helpful assistance in areas such as store administration and staff training. However, despite these optimistic claims, there has been little empirical validation.

Our paper seeks to help fill this void. Specifically, we aim to test the claims made by buying groups, and in particular whether they provide effective training for small stores' workforces. The question is of obvious importance to small retailers constrained in their human resource development efforts. Whereas traditional human resource management practices advocate thorough and systematic attention to the human resource function, the world of the small firm is typically so resource constrained (Welsh & White, 1981) that it cannot realistically attempt to implement such a comprehensive range of actions. What the small firm can do, though, is join a collective and receive assistance in workforce training and development. But is membership in a buying group related to higher levels of staff training and is that training linked to superior performance? Buying groups come in various shapes and sizes--are all buying groups' training programs equally effective or are there buying group-specific performance effect s?

In our paper we address these questions. We begin by taking a resource-based view (Barney, 1991) of the human resource management function. We then note the demonstrated tendency for smaller firms to undertake less comprehensive human resource activities. We continue by exploring the benefits offered by trade name franchise membership as a possible remedy for the small firm's shortcomings. Four hypotheses emanate from our discussion. Our first hypothesis concerns the relationship between employee training and firm performance, the second tests whether employee training is related to firm size, while the third focuses on the connection between buying group membership and performance. Building on the training-related advantages of the trade name franchising, we also posit an hypothesis of interaction between the variables of training and trade name franchise membership with performance.

THEORETICAL FRAMEWORK

Human Resource Management as a Source of Sustainable Competitive Advantage

The idea of a "human" resource finds its place in the broader perspective of strategic management known as the resource-based view of the firm (Barney, 1991; Mahoney, 1995; Mueller, 1996). According to this view, firms can be understood as collections of resources that permit differentiated offerings of products and services. To the extent the firm's offerings are judged by customers as sufficiently rare, valuable, inimitable, and nonsubstitutable (Barney, 1991), they allow the firm to extract superior rents and sustain future operations. At their most basic level, resources include physical, organizational, and human dimensions. Human resource management, therefore, is about creating and sustaining superior human resource contributions to firm offerings. Recently, Lado and Wilson defined the systematic activity of human resource management more comprehensively as "a set of distinct but interrelated activities, functions, and processes that are directed at attracting, developing, and maintaining (or disposin g of) a firm's human resources" (1994, p. 701). The management of a firm's human resources is important because human resources contribute to the formation of an organization's competencies, that is, "[the] firm-specific resources and capabilities that enable the organizations to develop, choose, and implement value-enhancing strategies" (Lado & Wilson, 1994, p. 702). By developing a distinctive set of firm-specific capabilities not easily copied or surpassed by rivals (Collis & Montgomery, 1995), an organization can lay claim to sustainable competitive advantage.

Human resource management is a multi-faceted process and includes several different elements: internal assessment, job design, interviewing, evaluation, selection, orientation, training, and coaching (DeSimone & Harris, 1998). Our interest in this article is with the specific component of training. Training is important to the firm because it equips employees with the technical and interpersonal skills essential for the organization to generate its product and service offerings. Training seeks to provide a variety of competencies including remedial, technical, and interpersonal. Likewise, training may be imparted through a diverse range of delivery modes: manual-based, in-house-trainers, customized external training, and general courses offered independently of the organization. Our focus in this study is not to explore issues of training content, but training delivery, and in particular how trade name franchises aid their members in imparting value-adding expertise to their employees. Therefore, our first h ypothesis is essentially confirmatory, namely that:

H1: Employee training is positively related to firm performance.

Human Resource Training and the Overlooked Small Firm

Despite the importance of the human resource function, not all enterprises appear equally able to realize the distinct advantages resident within their personnel; more specifically, smaller firms appear disadvantaged (Brown, Hamilton, & Medcoff, 1990). The rationale for this occurrence is multi-faceted. Compared to their larger counterparts, small firms are characterized by resource paucity (Welsh & White, 1981) and therefore have less slack to devote to training and development. Given their smaller size, any training initiatives are typically less efficiently offered, both in terms of costs associated with filling available spots (Vickerstaff, 1992) and in tailoring training content to firm-specific needs (Kirby, 1990). Complicating the mix even further is a general perception held by many employees in smaller organizations, namely, that their prospects for advancement are better realized in larger enterprises (Kruse, 1992; Morisette, 1993). When these actual and perceived impediments are considered alongsi de small firms' higher mortality rates (Storey, 1994), this finding, of training being positively related with firm size, is not overly surprising. Therefore, our second hypothesis is also confirmatory in nature, namely that:

H2: Employee training is positively related to firm size.

Trade Name Franchise Membership as a Small Firm Resource

There has been a significant increase in franchising research (Hoy & Shane, 1998). However, the vast majority of this work focuses on the synoptic mode of franchising known as business format franchising. In this form of franchising the franchisee essentially licenses the franchisor's business formula entire. In return, virtually all aspects of the franchisee's activities are predetermined. However, this is not the only form of franchising; another, less dominant form, trade name franchising, also exists. Compared to business franchising, trade name franchising (TNF) is much less extensive and typically involves a franchisee joining a franchise network, or buying group (Hardy & Magrath, 1987), in order to gain selected benefits including access to the network's trade name and procurement discounts through bulk buying. Each advantage finds support in a different theoretical rationale; advantages of identity emanate from the trade name serving as a signaling device for the franchisor brand equity (Wernerfelt, 1991), while advantages in procurement link to the minimization of transaction costs (Williamson, 1975). Taken together, these two advantages suggest that buying group membership should deliver significant advantage for its members. Hence:

H3: Buying group membership is positively related to firm performance.

Trade Name Franchise Membership as a Small Firm Training Resource

A third advantage of membership in the collective also warrants mention, however. Buying groups also aid their members by providing start-up and periodic ongoing training that facilitates the development of the store's workforce as a rare and valuable resource. At start-up, franchisors typically advise on such matters as store layout, advertising and promotion strategy, as well as instruction on in-store selling and point-of-sales inventory techniques. For established operations, training often includes updates in each of these same areas, but also includes information on new product line additions and instruction in how to provide store customers with specialized skill-intensive services. These start-up and ongoing franchisor contributions are of particular interest to our study insofar as the human resource training provided through trade name franchising suggests that franchisees can realize advantage not otherwise available. Furthermore, since franchisees are all linked into the same buying network, franchisors can tailor training around the common needs of all franchisees. By permitting franchisors to spread training costs over larger numbers than would otherwise be possible, they can also realize scale-based economies of instruction. Therefo re:

H4: Buying group membership strengthens the positive relationship between employee training and firm performance.

METHODOLOGY

Rationale for Industry Selection

We selected the retail hardware industry (SIC 5251) for this study for three reasons. First, the industry has a long history of fragmentation; as Thompson (1992) notes, the $100 billion do-it-yourself market, of which retail hardware constitutes but one segment, consists of tens of thousands of small, independent retailers. The second reason emanates from the recent evolutionary changes this industry has undergone; with the entry of several large chain players, the competitive dynamics for small incumbents are fundamentally altered (Do-it-yourself Retailing, 1994). Third, trade name franchising is a common occurrence in this industry; Hardy and Magrath (1987) estimate about half of all hardware stores are members of one or more buying groups.

Identification of Sample, Development of Research Instruments and Response Rates

We contacted 1,169 small hardware stores in the major U.S. metropolitan areas of Atlanta (GA), Chicago (IL), Kansas City (MO), Long Island (NY), Miami (FL), Minneapolis-St. Paul (MN), and San Diego (CA). Each store was asked to participate in a structured telephone interview and complete a mail-administered survey designed in accordance with Dillman's Total Design Method (Dillman, 1978). Both instruments were pre-tested on a separate group of ten buying group members in and around Winnipeg, Canada; all but one of the 20 sales and performance data points were subsequently validated by the members' Canadian headquarters, thus suggesting a high degree of validity. (In keeping with our promise of confidentiality, respondents' identities were at no time revealed to the head office.) Of the total sample, 340 (29.1%) were inaccessible (they had either ceased operations or were unavailable to answer the initial telephone call), 62 (5.3%) were incorrectly categorized as retail hardware stores, 110 (9.4%) were in oper ation and correctly classified as retail hardware stores but not interested in participating in our study. Follow-up calls were made to the 677 stores that were correctly classified as retail hardware stores and that were willing to participate in our study; 370 (31.6% of the total sample) did not return their survey, while 307 (26.3% of the total sample) did return their completed surveys. In terms of the total sample the 307 respondents represent 26.3%; in terms of the 787 stores deemed eligible to participate in the study, they represent 39.0%; in terms of the 677 stores that agreed to participate, the 307 responses represented an effective response of 45.3%.

Operationalization of Variables

Small retailer performance. Using scales developed with the assistance of a retail hardware industry expert (who identified appropriate increments for our sales and square-footage measures), we divided 1994 sales by store square footage to operationalize performance in terms of the retail industry's standard productivity measure of sales per square foot (Mason, Mayer, & Ezell, 1988). In order to more closely approach a normal distribution, we transformed the resulting product logarithmically.

Trade name franchise membership. Respondents were asked to indicate whether their store was a member of a trade name franchise network. In addition, we also asked respondents to provide the name of their franchise (e.g., Ace, True Value, etc.).

Employee training. We asked respondents how often employees attend technical training sessions, using a five-point Likert scale (1 = never, 2 = yearly, 3 = monthly, 4 = weekly, and five = daily).

Control variables. In recognition of the importance of munificence as a key environmental dimension (Castrogiovanni, 1991) we included control variables for both customer affluence (median per capital income) and competitor density (number of retail competitors identifying themselves as competing under the '52' SIC category). During the initial telephone interview we asked respondents to identify the zip codes comprising the critical mass of their store's trading area. We then collected data from the most recent U.S. census for each of the zip codes on each of the two dimensions and computed a measure of resident affluence and competitor density weighted by each zip code's population. We also asked respondents to provide information on their store's age (number of years in operation) and size (square footage).

ANALYSIS

Descriptive Profile of Sample and Correlation Matrix

The mean age of our respondents' stores was 38 years. The sample was almost equally divided between stores 5,000 square feet or less and those larger than 5,000 square feet. Just over half (51.8%) reported receiving training on an annual basis, while almost one-third (31.3%) reported receiving no technical training. By comparison, a small minority (14.9%) reported receiving training on at least a monthly basis. Over 80% (250 stores) of the sample reported belonging to a trade name franchise, with nineteen different trade name franchise networks identified; the most common affiliations (identified by at least 10 respondents) were Ace (74 stores), True Value (72 stores), Servistar (32 stores), Sentry (18 stores), and United (12 stores). In carrying out our hypothesis testing we employed multiple regression analysis; a matrix of the variables used is found in Table 1; in Table 2 we report results from our set of regression models testing our study's four hypotheses.

H1: Is training related positively to firm performance? As Model 1 reports, our results uphold the general proposition Linking employee training and superior firm performance. Also noteworthy is a negative relationship between firm size and performance; this is attributable in large part to the size-sensitive nature of the performance operationalization utilized.

H2: Is training related positively to firm size? As Model 2 demonstrates, workforces of smaller firms are less likely to receive training, thus confirming our second hypothesis. An interesting concomitant finding concerns the relationship between training and competitor density; this suggests that the training is seen as more important in more competitive environs.

H3: Is trade name franchise membership related positively to firm performance? Results in Model 3 support an affirmative answer as performance is significantly enhanced by TNF membership. Also noteworthy was a trend relating competitor density to firm performance; however, no parallel pattern was apparent in any of the models between resident affluence and performance.

H4: Does trade name franchise membership enhance the hypothesized positive relationship between training and firm performance? Model 4 shows each of the primary independent variables continuing to have a positive relationship with firm performance; however, their interaction variable tells a less straightforward story. In order to better understand the nature of this interaction we stratified our sample by level of training (firms receiving either no or only yearly training were put in a 'Low' training group while firms receiving either monthly or weekly training were placed in a 'High' training group). We then combined the High/Low training dimension with TNF Membership/Independent status to generate a four-cell matrix. A Chi-square test of association revealed no significant relationship (p = 0.889); however, analysis of variance on performance levels across the four cells revealed an interesting pattern (Table 3). While the mean performance of buying group members was higher than that of independents, thi s pattern was not sustained in the face of independents reporting high levels of training. Analysis of tetrad differences across the four cells clarified this difference; performance of High/Independent retailers was significantly higher than any of the three other cells with no other contrasts achieving significance.

Exploring Trade Name Franchise-specific Effects

In our introduction we posited the possibility of differences in training across franchise networks. In this final section we explore this possibility. First, we ran regression models for the most frequently mentioned TNFs (that is, Ace, True Value, and Servistar). In Table 4 we report results from franchise-specific regression models that, owing to smaller sample sizes, included only firm size and frequency of training as predictors. Significant differences across the three models are apparent; the most notable contrast is the nonsignificant relationship between training and performance for Ace members and the positive relationship reported by both True Value and Servistar members. This provides early support for our exploratory proposition of interfranchise differences.

DISCUSSION AND CONCLUSIONS

The findings of this study aid in better understanding the nature of the small firm's human resource training challenge and in particular the extent to which the collective affiliation known as the trade name franchise helps in addressing this challenge. Though researchers have devoted significant energies to understanding the effectiveness of the franchise as a governance structure, there has been little work on studying its effectiveness as a training mechanism. In our study, we attempt to fill this gap by exploring the extent to which membership in a trade name franchise assists small firms in realizing sustainable competitive advantage. Our findings provide coarse-grained support for the two freestanding hypotheses of training and trade name franchise membership. However, our analysis also suggests that the contributions of franchise membership do not spill over into staff training. This finding suggests that imparting skill-based competence to small retailers' workforces may be more problematic than at first understood. Our findings also point to the possibility of effective small firm training being a firm-specific phenomenon; this is supported most clearly by the performance levels of the independents reporting high levels of training. Also compounding our interpretation is the concomitant finding of TNF training's relationship with performance being franchise-specific; simply stated, some TNFs appear more effective in imparting performance-enhancing skills than others. At the very least, this suggests that the choice of trade name franchise network is a nontrivial decision for the small and medium-sized retailer.

Several caveats not only footnote these findings but also suggest avenues for further research. First, we have glossed over inter-firm differences in product and service mix; to what extent is training's contribution to superior performance leveraged more easily for some product or service mixes than others? Second, we have only studied one market segment: retail hardware. However, buying groups are found in several different sectors (Hardy & Magrath, 1987). To what extent are the findings reported here supported in other industry contexts? A third issue arises from performance differences observed across buying groups. Since, as Delaney and Huselid note, human resource literature is marked by the dubious distinction of "virtually no two studies [measuring] HRM practices in the same way" (1996, P. 967), further work is necessary on how different franchise groups operationalize "training." Likewise, further research on the effects of greater versus lesser dependence by trade name franchisees on franchisors is warranted; in post hoc analysis on the percentage of store merchandise procured through their respective trade name franchisor, members of the Ace and True Value networks reported acquiring significantly higher amounts (85.1 and 79.6%, respectively) through their franchisor, compared to Servistar franchisees, who secured less than two-thirds (63.6%) of their inventory. Given the buying group-specific findings reported earlier, an ominous question is hinted at--could excessive dependence on a trade name franchisor essentially neutralize the franchisee's capacity for entrepreneurial initiative?

Finally, given the outstanding performance reported by the sub-group of independents, we may need to ask an even more basic question: can the franchise form of organization contribute significantly and consistently to the formation of customer-perceived value? This concern arises from Mahoney's (1995) proposition of invisible resources emanating from three types of information flow: environmental information (such as learning customer preferences), corporate information (relating to brand name and reputation), and internal information (resulting in increased information-processing capacity). Since as Mueller notes, "strategic assets may be largely the unintended result of many patterns of informal and unofficial behavior" (1996, p. 774, emphasis added), might trade name franchisors be inherently limited to only coarse-grained corporate information flows, and hence excluded from facilitating finer-grained environmental and internal information flows?

In summary, many small retail establishments face the onslaught of giant scale-advantaged players invading their competitive domains. These big box competitors are able to gain scale-related advantage in many different facets of operation including procurement, advertising, and employee training. This trend prompts many industry observers to wonder whether the days of the small retailer are numbered. However, another more hopeful perspective also exists. If small firms band together, this line of thinking goes, they can collectively offset some of the giant player's advantage. Our research provides some, albeit limited, support for this hypothesis. While membership in a collective appears to enhance members' performance, this enhancement does not appear to manifest itself in the trade name franchisor's training offerings.

Reginald A. Litz is Associate Professor of Management in the I. H. Asper School of Business at the University of Manitoba.

Alice C. Stewart is Director of Strategic Analysis and Planning and an Assistant Professor of Strategic Management at Fisher College of Business at Ohio State University.

The authors gratefully acknowledge the funding provided by the United States Small Business Administration--Office of Advocacy (Contract #SBA-8124-0A-94) and the University of Pittsburgh 1994-95 Central Research Development Fund (Grant #5-31883) for this research project.

REFERENCES

Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99-120.

Brown, C., Hamilton, J., & Medcoff, J. (1990). Employers large and small. Cambridge, MA: Harvard University Press.

Castrogiovanni, G. J. (1991). Environmental munificence: A theoretical assessment. Academy of Management Review, 16(3), 542-565.

Collis, D., & Montgomery, C. (1995). Competing on resources: Strategy in the 1990s. Harvard Business Review, 71(July-August), 118-128.

Delaney, J., & Huselid, M. (1996). The impact of human resource management practices on the perceptions of organizational performance. Academy of Management Journal, 39(4), 949-969.

DeSimone, R. L., & Harris, D. M. (1998). Human resource development. Fort Worth, TX: Dryden Press.

Diliman, D. A. (1978). Mail and telephone surveys: The total design method. New York: John Wiley & Sons.

Do-it-yourself Retailing. (January 1994). Competing head-to-head: In their own words, hardware and home improvement retailers share their competitive strategies. p. 46-54, 69.

Ehrenfeld, T. (1995). The demise of mom and pop? Inc., 17(1), 46-48.

Hardy, K., & Magrath, A. (1987). Buying groups: Clout for small businesses. Harvard Business Review, 63 (September-October), 16-18, 20, 22, 24.

Hoy, F., & Shane, S. (1998). Franchising as an entrepreneurial venture form. Journal of Business Venturing, 13(2), 91-94.

Kirby, D. A. (1990). Management education and small business development: An exploratory study of small firms in the UK. Journal of Small Business Management, 28(4), 78-87.

Kruse, D. (1992). Supervision, working conditions and the employer size-wage effect. Industrial Relations, 31, 229-249.

Lado, A., & Wilson, M. (1994). Human resource systems and sustained competitive advantage: A competency-based advantage. Academy of Management Review, 19(4), 699-727.

Mahoney, J. (1995). The management of resources and the resource of management. Journal of Business Research, 33(1), 91-101.

Mason, J. B., Mayer, M. L., & Ezell, H. F. (1988). Retailing (third edition). Piano, TX: Business Publications, Inc.

Miller, C. (1992). Big chains battle for market share in home improvement. Marketing News, 26(20), 1, 10-11.

Morisette, R. (1993). Canadian jobs and firm size: Do small firms pay less. Canadian Journal of Economics, 26, 159-174.

Mueller, F. (1996). Human resources as strategic assets: An evolutionary resource-based theory. Journal of Management Studies, 33(6), 757-785.

National Home Center News. (December 14, 1992). Elmont store wreaks havoc on local outlets: Survivors ponder future as Home Depot expands on Long Island, N.Y. p. 160, 162, 164.

Storey, D. J. (1994). Understanding the small business sector. London: Routledge.

Thompson, R. (1992). There's no place like Home Depot. Nation's Business, 80(2), 30-33.

Vickerstaff, S. (1992). The training needs of small firms. Human Resource Management Journal, 2, 1-15.

Welsh, J. A., & White, J. F. (1981). A small business is not a little big business. Harvard Business Review, 59(4), 18-32.

Wernerfelt, B. (1991). Brand loyalty and market equilibrium. Marketing Science, 10(3), 229-245.

Williamson, O. (1975). Markets and hierarchies: Analysis and antitrust implications. New York: Free Press.

                      Correlation Matrix of Variables
                   Mean
                   (s.d.)   [1]         [2]    [3]       [4]    [5]
[1] Performance     0.202  1.000
                   (0.16)
[2] Median Income 37745    0.039        1.000
                  (12465)
[3] Competition    12.9    0.107 [+]   -0.015 1.000
                    9.1
[4] Firm age       37.5    0.072       -0.016 0.003      1.000
                  (24.4)
[5] Firm size       2.34  -0.377 [***]  0.070 0.010     -0.064 1.000
                   (0.70)
[6] Training        1.84   0.124 [*]    0.013 0.119 [*]  0.011 0.222 [***]
                   (0.71)
[7] TN-Franchise    0.81   0.008        0.076 0.021      0.021 0.341 [***]
                   (0.38)
                   [6]        [7]
[1] Performance
[2] Median Income
[3] Competition
[4] Firm age
[5] Firm size
[6] Training      1.000
[7] TN-Franchise  0.173 [**] 1.000
(+.)p [less than] .10,
(*.)p [less than] .05,
(**.)p [less than] .01,
(***.)p [less than] .001
                      Summary of Regression Analysis
                      (Dependent variable for Models
                        1, 3, and 4: Performance as
                           sales per square foot
                      Dependent variable for Model 2:
                      Frequency of employee training
                       (Standardized Betas reported
                           with t-ratio below))
                                 Model 1:         Model 2:
                                Effects of       Effects of
                                Training on     Firm Size on
Variable                        Performance       Training
Control variables
 Median Income                     0.068           -0.011
                                  (1.275)         (-0.187)
 Competitor density                0.087            0.115
                                  (1.621)          (2.003) [*]
 Age of store                      0.044            0.022
                                  (0.818)          (0.380)
 Size of store                    -0.425            0.239
                                 (-7.709) [***]    (4.155) [***]
Independent variables
 Frequency of training             0.206
                                  (3.727) [***]
 TNF Member (1 = yes: 0 = no)
 Freq. of training x TNF Member
Constant                           0.281            1.169
                                  (6.015) [***]    (5.854) [***]
[R.sup.2]                          0.201            0.071
Adjusted [R.sup.2]                 0.186            0.058
df                               279              284
F                                 14.016 [***]      5.396 [***]
                                 Model 3:          Model 4:
                                  Effects         Effects of
                                 of TNF on      Training x TNF
Variable                        Performance     on Performance
Control variables
 Median Income                     0.061            0.059
                                  (1.126)          (1.113)
 Competitor density                0.101            0.082
                                  (1.878) [+]      (1.543)
 Age of store                      0.040            0.054
                                  (0.733)          (1.019)
 Size of store                    -0.436           -0.450
                                 (-7.551) [***]   (-7.826) [***]
Independent variables
 Frequency of training                              0.434
                                                   (3.907) [***]
 TNF Member (1 = Yes: 0 = no)      0.140            0.411
                                  (2.421) [*]      (3.215) [**]
 Freq. of training x TNF Member                    -0.431
                                                  (-2.489) [*]
Constant                           0.329            0.169
                                  (7.356) [***]    (2.779) [**]
[R.sup.2]                          0.181            0.232
Adjusted [R.sup.2]                 0.167            0.212
df                               282              227
F                                 12.503 [***]     11.930 [***]
(+.)p [less than],
(*.)p [less than].01,
(***.)p [less than] .001
                  ANOVA of Training Level and Trade Name
                           Franchise Membership
                 (Dependent variable: Firm performance as
                          sales per square foot)
Source        df Sums of Squares Mean Square F-ratio             Prob
Constant       1     11.4159       11.4159   466.11     [less than or
                                                      equal to]0.0001
TNF            1      0.084323      0.084323   3.4429          0.0646
 (Yes/No)
Training       1      0.257702      0.257702  10.522           0.0013
  (High/Low)
Interaction    1      0.162281      0.162281   6.6260          0.0106
Error        286      7.00464       0.024492
Total        289      7.26713
                      Summary of Regression Analysis
         (Dependent variable: Performance as sales per square foot
             (Standardized Betas reported with t-ratio below))
                                   Model 5         Model 6        Model 7
                                   Ace TNFs    True Value TNFs Servistar TNFs
Internal control variable
 Size of store (square footage)  -0.629          -0.291        -0.732
                                (-6.800 [***])  (-2.449 [*])   -5.873 [***])
Independent variables
 Frequency of training            0.098           0.233         0.294
                                 (1.054)         (1.961 [+])   (2.358 [*])
                                  0.488           0.262         0.495
Constant                         (9.037 [***])   (3.028 [**])  (6.202 [***])
[R.sup.2]                         0.402           0.110         0.548
Adjusted [R.sup.2]                0.385           0.084         0.518
df                               70              66            30
F ratio                          23.503 [***]     4.098 [*]    18.188 [***]
(+.)p [less than].10, (*.)p [less than].05,
(**.)p [less than].01, (***.)p [less than].001