The Dynamics of Increasing Internship Conversion Rates:
Practical Implications for Retail-related Businesses

Turnover in the retailing industry is traditionally high and since the US typically has a fairly abundant young labor force, retailers experienced high turnover rates with minimal consequences, until now (National Retail Federation [NRF], 2008). Experts (Bureau of Labor Statistics [BLS], 2006; Employment Policy Foundation, 2005) are predicting a labor shortage, especially in the retailing, hospitality and food services industries, due to the demographic nature of their employees (i.e. younger and first-time employees) (BLS, 2007). Nevertheless, as the pool for entry-level employees shrinks, and the job market continues to get more and more competitive (Pianko,1996; US Department of Labor, 2003), retailers will need to revisit their current attraction/retention strategies quickly if they expect to effectively respond to this upcoming labor shortage (Retensa, 2008).

One potentially viable strategy available to retailers is to offer internship opportunities to college students (i.e., potential future hires).Companies create awareness about their organizations by offering internships, and internships offer college students the opportunity to work with professionals in their field and experience the industry firsthand (Coco, 2000; Henry, Rehwaldt & Vineyard, 2001; Lampe & Rothman, 2004).

Internships offer students the opportunity to participate in a realistic job experience. Internships also give the company a chance to evaluate students’ work capabilities before a potential offer for fulltime employment is extended. In addition, internships allow both the intern and the company the opportunity to determine if the intern possesses the qualities necessary to adapt to the firm’s culture. Converting good interns (i.e., those that have received a job offer) to full-time employees is becoming the preferred path to a permanent position within the firm (Gardner, 2004); however, conversion rates within the retailing industry, particularly at the entry-level store management level, are lower than employers would like (Pedersen, 2007; Research and Markets, 2006). According to Pedersen (2007), conversion rates are typically calculated by dividing the number of accepted offers by the number of total offers. For companies to stay competitive in today’s job market, a conversion rate of 50- 70% is the target benchmark (Pedersen, 2007).

Increasing internship conversion has been shown to be beneficial to the employer for a number of reasons, specifically: 1) interns represent a partially trained workforce that immediately contributes to the organization (Dixon, Cunningham, Sagas, Turner & Kent, 2005), 2) it saves a significant amount of money both in hiring and training costs (Nielsen & Porter, 1983; Pianko, 1996), and 3) it potentially augments job satisfaction and organizational commitment and decreases voluntary turnover. According to Gault, Redington and Schlager (2000) interns typically experience higher job satisfaction than non-interns; therefore, since research shows job satisfaction and organizational commitment to be significant antecedents of turnover intentions (Firth, Mellor, Moore & Loquet, 2004; Meyer, Stanley, Herscovitch & TopoInytsky, 2002) it is plausible that increasing internship conversion may contribute to higher levels of job satisfaction and organizational commitment and decreased voluntary turnover.

Given that interns represent a potentially viable pool of new hires (Sessions, 2006), and conversion rates remain relatively low (Pedersen, 2007; Research & Markets, 2006), understanding the factors that can increase conversion rates would be constructive contribution to retail practice. Hence, our research focuses on college students who successfully completed a business or retail-related internship. Variables that may increase conversion rates include: psychological contract outcomes (i.e., fulfilled/ unfulfilled obligations), supervisory support, job satisfaction, and organizational commitment (affective and continuance). Understanding these relationships is important for educators as they prepare students for careers and help them form their career expectations. In addition, we may begin to understand why some interns choose not to accept a job offer for full-time employment from the company after graduation (Knight, Crutsinger & Kim, 2006; Rothman, 2007).

Review of Relevant Literature and Hypothesis

The perceived value of collegiate internships from both the employers’ and students’ point of view is illustrated by the significant growth in student/employer involvement over the past two decades (Cook, Parker & Pettijohn, 2004). But, the needs and objectives of the interns must be satisfied for such programs to persist (Lampe & Rothman, 2004). In a recent study regarding employers’ perceptions of the benefits of internship programs, Metzger (2004) found 97.4% of employers felt that students gain marketable skills from participating in internship programs, 96.9% agreed that internship experiences assist in students’ career development, 75.7% stated that internship programs are cost-effective mechanisms for recruitment, and 58.8% felt that hiring previous interns provides for lower long-term employee attrition.

Because the internship experience, albeit short and temporary in nature, represents an integral time for forming impressions of, or commitment to, the organization (Dixon, et. al., 2005), many organizations have high expectations of hiring for permanent positions from their intern pool (Gault et. al., 2000; Sessions, 2006). Hence, to assist management in gaining more control over the salient motivators that cause interns to turn down an offer and to augment motivators that cause interns to accept an offer upon internship completion, we examine the aforementioned variables (psychological contract outcomes and supervisory support), in addition to more commonly researched variables (e.g., organizational commitment and job satisfaction).

Psychological Contract Outcomes(i.e., Fullfilled or Unfulfilled Obligations)

Psychological contracts are individuals’ beliefs in mutual obligations between themselves and other parties, such as an employer or supervisor (Rousseau, 2000). Recruitment experiences create understandings regarding the promises workers and employers make to each other. Since psychological contracts of new hires are only rudimentary, expectations and obligations are continually modified during the socialization process (Thomas & Anderson, 1998).

However, if new hires are unable to revise their psychological contracts in a way that allows them to properly adjust and socialize to the organizational culture, and meet their own and their employers’ beliefs and expectations, a psychological contract breach occurs; which often leads to turnover (Robinson & Rousseau, 1994). Since the highest level of voluntary turnover occurs within the first 12 months on the job (BLS, 2006), it is important to understand how and why newcomers adjust their psychological contracts over time and how it affects job satisfaction, affective organizational commitment and conversion intentions (see Figure 1).

Supervisory Support

Because psychological contract implementation is carried out predominantly by the supervisor and the supervisor observes the employee’s compliance with the company’s psychological expectations of the employee, the supervisor becomes a key player in the recruitment/retention strategy (Gentry, Kuhnert, Mondore & Page, 2007; Firth et. al., 2004). The number one reason why employees leave or thrive in an organization is not their specific job, the organization, or their salary, but rather, their manager (“Good Managers,” 2006; Jamrog, 2002).

It is intuitively obvious that a structured internship program with an adequate amount of supervisory support or mentoring can help promote positive work habits of the intern. However, one criticism of internship programs is the inconsistency of supervision and organization of tasks (Gault et. al., 2000). Compared to full-time employees, interns may face less clearly defined lines of authority and less contact with supervisors (Dixon et. al., 2005); which may be due to the temporary nature of the internship program (Coco, 2000). If organizations structure the supervision of interns in a way that maximizes interns’ experience with the organization, it should enhance their sense of loyalty and/or belonging to the organization (i.e., affective organizational commitment).

Additionally, Firth et al. (2004) found that supervisory support mediated the relationship between job stressors and job satisfaction, commitment to the organization, and intention to quit for retail salespeople. Since newcomers tend to rely on their supervisor to carry out many of the contractual obligations owed to them, employees are likely to view their supervisor as the chief agent for establishing and maintaining their psychological contracts. Thus, we suggest the following hypothesis:
H1: Interns’ supervisory support will have a positive effect on psychological contract outcomes of employer obligations.

Job Satisfaction, Organizational Commitment, and Conversion Intentions

Job satisfaction is defined as a person’s overall feeling about work and the work organization (Champoux, 2003). Even though many factors affect a person’s feelings about work and the organization, the job is a basic connection between the person and the employing organization. Therefore, a person’s job has some effect on their feelings of overall job satisfaction. Job satisfaction has been shown to have a direct and negative impact on intent to leave (Netemeyer, Burton & Johnston, 1995; Igbaria & Guimaraes, 1993; Tate, Whatley & Clugston, 1997), and an indirect affect on intent to leave that is mediated by organizational commitment (Netemeyer et. al., 1995; Schaubroek, Cotton & Jennings, 1989).

Organizational commitment is defined as “the relative strength of an individual’s involvement in a particular organization” (Steers, 1977, p. 46) and it is a significant predictor of turnover and/or intention to stay/leave (Mowday, Steers & Porter, 1979; Steers, 1977). In our study, conversion intentions assume the same role as intentions to stay/leave a company; upon receiving an offer from the internship company, the intern either intends to remain and start a career with the company or leave to pursue other opportunities. Meyer and Allen (1991) identified three dimensions of organizational commitment, affective, continuance, and normative; we include only affective and continuance commitment in our study.

Affective commitment is an attitudinal process where people eventually think about their relationship with the organization in terms of value and goal congruency. People with strong affective commitment remain with the organization because they want to (Clugston, 2000).

Continuance commitment is a person’s need to remain with the organization based on the costs associated with leaving (Meyer & Allen, 1991), and it is apparent in two major ways: 1) as individuals gain tenure, they accrue investments (i.e., pension plans, seniority, etc.) which may be sacrificed or lost by changing jobs; and 2) individuals may feel like they have to remain in their job because they have no alternative job prospects. Since continuance commitment develops as employees make investments that would be lost or reduced in value if they left the organization, it is expected to be unrelated or negatively associated to job satisfaction and job performance (Meyer, Allen & Smith, 1993). Thus, we pose the following hypotheses:
H2: : Interns’ supervisory support will have a positive effect on a) job satisfaction and b) affective organizational commitment; and
H3:Interns’ perceptions of the extent to which psychological contract outcomes of employer obligations were fulfilled will have a positive effect on a) job satisfaction, b) affective organizational commitment and c) conversion intentions.

In our study, the final outcome variable, conversion intentions, consists of the interns’ intent to accept an offer. This indicates their decision to remain with the organization and begin full-time employment upon graduation. According to Allen, Drevs and Ruhe (1999), committed employees wish to remain with their employing organization. Additionally, research has consistently shown a negative relationship between commitment and turnover (Cohen, 1993; Cotton & Tuttle, 1986; Mathieu & Zajac, 1990). Also, commitment to the organization has been repeatedly found to be positively influenced by job satisfaction (Cotton & Tuttle, 1986; DeConinck & Bachmann, 2005). Moreover, a preponderance of research that includes both job satisfaction and organizational commitment indicates that job satisfaction is an indirect predictor of turnover through organizational commitment, primarily because commitment is a more global and stable evaluative linkage between the employee and the organization (Johnston, Parasuraman, Futrell & Black, 1990; Meyer & Allen, 1997). Hence, we pose the following hypotheses:
H4:Interns’ job satisfaction will have a positive effect on a) affective organizational commitment and a negative effect on b) continuance organizational commitment; and
H5:Interns’ a) affective, and b) continuance organizational commitment will have a positive effect on conversion intentions.

Research Methods

Proposed Path Model and Theoretical Framework

We developed a proposed path model using theoretical foundation from psychological contract theory (Rousseau, 2000), and organizational socialization theory (Van Maanen, 1975).

Psychological contract theory combines perceptions of employer-based obligations with employee obligations, and suggests outcomes related to job satisfaction, organizational commitment and turnover intentions (Robinson, Kraatz & Rousseau, 1994). Organizational socialization theory explains how newcomers learn the culture and values of new job settings; socialization depends on supervisory support to enable the socialization process (Chao, O’LearyKelly, Wolf, Klein & Gardner, 1994; Ostroff & Kozlowski, 1992; Van Maanen, 1975).

The model proposes the relationships between psychological contract outcomes of employer obligations, supervisory support, job satisfaction, organizational commitment (affective & continuance), and conversion intentions. The model (see Figure 1 - Appendix A) is intended for use with college students who have completed a business or retail-related internship.

Sample Participants

A total of 250 college student interns were recruited from participating companies/universities in the United States. The population consists of college students who completed a business or retail-related internship in the USA during 2006 or spring 2007. Only interns who worked 26 or more hours per week were included. About 60% of the interns surveyed are female, and 31% are male. The average age is 22 years, and a majority of the interns are juniors or seniors (73%). Interns completed internships at a variety of business or retail-related firms throughout the USA (i.e., retailers, banks, marketing firms, etc.).

Pretest

Stage one of the instrument pre-test involved two separate focus-group sessions with six interns each (N=12). This allowed us to test the content validity and item relevancy of the survey instrument and indicated items that needed to be modified. For example, the original employer/employee obligations (psychological contract outcomes) scales include items such as: ‘Long term job security’, ‘Willingness to accept a transfer’, and ‘Spending a minimum of two years with the organization’ (Robinson et. al., 1994; Rousseau, 1990). Focus group participants felt that these items were not relevant to their internship experience, and should be replaced with ‘Anticipated number of working hours was approximately equal to the number of hours actually worked’, and ‘Willingness to do a variety of job tasks’. After making these modifications , we collected data for stage two of the pre-test by sending a cover letter (via e-mail) explaining the study, along with an HTML link to the on-line survey to a subset of 75 college juniors and seniors who had completed a business or retailrelated internship at a mid-western university. We made minor wording changes as indicated from the pretest and omitted seven items as a result of the pretest and reliability checks (i.e., Cronbach alpha).

Procedure

To attain the necessary sample size, we utilized two different methods for data collection: 1) Partnership with companies (HR representative forwarded cover letter and HTML link to the on-line survey to interns), and 2) Cooperation with faculty from five universities who had access to internship students (faculty member forwarded cover letter and HTML link to the on-line survey to interns).

Instrument and Measures

To measure the constructs in the proposed path model (see Figure 1 - Appendix A), we used an online, self-administered, questionnaire developed from existing scales (see Table 1 - Appendix B for a full list of scale items).

Psychological Contract Outcomes: Employer Obligations

Supervisory Support

Job Satisfaction

Organizational Commitment

Conversion Intentions

Four items measure interns’ intent to accept an offer from the organization (Robinson, 1996, with reported alpha coefficient of 0.86). The response format is from 1 (Strongly Disagree) to 7 (Strongly Agree).

Data Analysis and Results

Structural Equation Model

Data were analyzed using structural equation modeling/EQS 6.1 (Bentler, 2004). Consistent with the guidelines proposed by Anderson and Gerbing (1982), we first estimated the measurement model and continued with the structural model. We estimated the measurement model using a multi-step process of confirmatory factor analysis (CFA) (Kaplan, 2000).

Confirmatory Factor Anlaysis

To assess dimensionality, individual constructs were subjected to confirmatory factor analysis (CFA). The aim of the analysis was to determine unidimensionality of the constructs and provide partial assessments of model fit. All measurement items from the survey were entered into the analysis by respective construct. Standard procedures were used for model assessment. First, fit statistics were checked to evaluate model fit. The chi-square statistic is an absolute measure of model fit; however, it is biased when dealing with large samples (200 or >), complex models, and models with large numbers of indicators (Kline, 2005). Since these conditions apply to the present model, an alternative fit index of absolute fit is also referenced, such as Root Mean Square Error of Approximation (RMSEA). RMSEA is a particularly meaningful index of absolute fit that measures how well the model would fit the population covariance matrix (if available). RMSEA values less than .05 indicate good fit, while values ranging from .08 to .10 indicate moderately acceptable fit (Bollen & Curran, 2006; Browne & Cudeck, 1993).

In addition to absolute indices of fit, incremental fit indices are also considered. Incremental fit indices compare the hypothesized model to the null model, and generally include the Bentler-Bonnett Normed-Fit Index (NNFI) and the Comparative Fit Index (CFI). Perfect fit for all incremental fit indices is 1.0 (Kline, 2005). Also considered when assessing model fit is the Standardized Root Mean Square Residual (SRMR). The SRMR transforms both the sample covariance matrix and the predicted covariance matrix into correlation matrices; thus, it is the overall difference between the observed and predicted correlations. SRMR values less than .10 are generally considered favorable (Kline, 2005).

Second, item to factor loadings were evaluated to determine convergent validity of the model. Item to factor loadings that are positive and significant support the convergent validity of the model. EQS modification indices (i.e., standardized residuals) were also checked for item to factor cross-loading to determine discriminant validity. Cross-loaded items that produced high error estimates were subjected to further evaluation by checking their standardized residuals. Standardized residuals are considered problematic if they form a pattern of error among construct indicators (Bentler, 2004; Bollen, 1989). To provide an assessment of construct validity, crossloaded items with high error terms (ζ), and low standardized estimates (λ’s) were removed one at a time, assessing model fit after each revision.

Items retained and deleted per construct as a result of model building are noted in Table 1, Appendix B. Acceptable fit was attained for the SS construct, thus no items were deleted. However, due to high construct cross-loadings and low standardized estimates (λ), items V1, V2, V4 and V8 were removed from the PCO_ER construct, which resulted in improved model fit. Additional deletions due to high error terms (ζ) and/or low standardized estimates (λ), included items: V26, V28, V30, and V32 from the JS_job construct, V43 from the AOC construct, V49, V50, and V51 from the COC construct and V57 and V60 from the CI_A construct.

For a final assessment of discriminant validity, we looked at the correlations between each construct and calculated the square root of the variance extracted (AVE) for each construct (seen on the Diagonal in Table 2). Assessment of construct correlations revealed a high correlation between PCO_ER and JS_job (r = .819) and PCO_ER and AOC (r = .705). Thus, it is possible that this lack of discriminant validity between these constructs may suppress the path coefficients of certain hypothesized relationships and negatively impact model fit. However, all of the constructs meet or exceed the desired AVE value of .70 or greater (Fornell & Larcker, 1981), except for CI_A (sq. root AVE = .662); which could be due to its moderately high correlation with AOC (r = .667). Logically, this high correlation is expected since AOC measures the level of attachment, which is linked to staying with the company (i.e., conversion intentions). Furthermore, according to Neter, Wasserman, and Kutner’s(1990) suggested variance inflation factor (VIF) criterion, correlations less than .92 will not cause serious multicollinearity problems in regression-based models. Thus, no other modifications were made (see Table 2 - Appendix B).

To assess the internal consistency of the revised construct scales, we used alpha coefficients and variance extracted (see Table 3 - Appendix D). Scales for all six constructs meet or exceed minimum levels (.70) of acceptable reliability (Nunnally, 1978). Variance extracted also meets or exceeds the minimum standard of .50 for all constructs except CI_A (variance extracted = .438) (Hair, Anderson, Tatham & Black, 1998). Perhaps the low variance extracted for the CI_A measure is because it was the most difficult scale to modify from a full-time employee’s perception of intent to remain to an intern’s perception of intent to accept an offer from the internship company (see Table 3 - Appendix C).

CFA revealed a significant chi-square statistic of 1023.92 (df = 446, p ≤ .001). The Root Mean Square Error of Approximation (RMSEA) of .072, combined with the following incremental fit indices indicate moderately acceptable fit: Non-Normed Fit Index (NNFI) = .879, Comparative Fit Index (CFI) = .891, and Standardized Root Mean-Square Residual (SRMR) = .063. CFA results were not as desirable as preferred; therefore, we investigated the standardized solution and standardized residuals. Final review of the CFA’s standardized estimates (λ’s) revealed significant (p < .05) loadings for each variable of .50 or higher (Bentler, 2004) and low standardized residuals. Given that the loadings were satisfactory and the largest standardized residual was fairly low, no further modifications were made.

The next step involved estimation of structural parameters and hypothesis testing (see Figure 1). Factor covariances of the six model constructs (i.e., covariance matrix) produced by the measurement model served as the input data for the structural model. The analysis produced a non-significant chi-square statistic (x ²) = 10.78, 5 df, p > .056) with a RMSEA of .068, indicating moderate model fit. In addition, the following incremental fit indices also indicate reasonable fit: Non-Normed Fit Index (NNFI) = .972 and Comparative Fit Index (CFI) = .993 with a Standardized Mean-Square Residual (SRMR) of .041. Strong and distinct item-factor loadings resulted for all model dimensions.

Hypothesis Testing

Hypothesis 1 posits that SS will have a positive effect on PCO_ER, which is supported (γ = .451, p < .05) and consistent with Robinson and Rousseau (1994), who state that perceptions of fulfilled psychological contract obligations are dependent on a contract formed under specific conditions that are influenced by ongoing interactions between the employee and organizational representatives, such as a supervisor.

Hypothesis 2a posits that supervisory support (SS) will have a positive impact on job satisfaction with the job/work itself (JS_job). A positive but nonsignificant relationship (γ = .030, p > .05) implies that interns in the current study do not feel that supervisory support enhances their job satisfaction as strongly as previous evidence suggests for full-time employees (Babin & Boles, 1996; Jamrog, 2002). Perhaps, due to the short-term nature and structure of the internship (i.e., moving from one functional area to another), interns in the current study required more guidance, leadership and support (i.e., mentoring/coaching) from their supervisors regarding the job/work itself than they actually received. For H2b, positive and significant support is found for the relationship between SS and affective organizational commitment (γ = .260, p < .05), signifying that supervisory support helps interns feel a sense of belonging to the company (Clugston, 2000).

Interns’ perceptions of the extent to which employer obligations (PCO_ER) were fulfilled is hypothesized to have a positive effect on job satisfaction with the job (H3a), affective organizational commitment (H3b), and conversion intentions (H3c). A positive and significant relationship exists between PCO_ER and JS_job (β = .690, p <.05), supporting H3a; when interns feel that their internship employer fulfills the expected obligations they are more satisfied with their job. A positive and significant relationship is indicated between PCO_ER and affective organizational commitment (β = .647, p < .05), thus supporting H3b; when interns feel that their internship employer fulfills their expected obligations, they will remain with their internship company because they want to (e.g., affective organizational commitment); and this trust in a reciprocal relationship enhances their sense of belonging and loyalty to the company. This is consistent with Meyer and Allen’s (1991) findings. We also find a significant and positive relationship between PCO_ER and conversion intentions (β = .283, p < .05), supporting H3c.

Hypothesis 4a proposes that interns’ job satisfaction with the job (JS_job) will have a positive effect on affective organizational commitment (AOC). This hypothesis is supported (β = .694, p < .05), suggesting that interns’ satisfaction with the job/work itself significantly contributes to their strong sense of attachment with the organization (AOC) because their internship has proven to be a satisfying experience (Meyer et al., 1993; Caykoylu, Egri & Havlovic, 2007).

In hypothesis 4b, we propose that job satisfaction with the job (JS_job) will have a negative effect on continuance organizational commitment (COC). Interestingly, a significant, positive relationship is indeed found (β = .295, p > .05); the more satisfied the intern, the more likely the intern would be committed to the organization due to the costs of leaving. Perhaps the current competitive job environment and the weak economic climate influenced the satisfied interns to feel more committed to a “sure thing.”

For the last set of hypotheses, we propose that AOC (H5a) and COC (H5b) will have a positive effect on conversion intentions. Positive and significant results are found for both hypotheses (β = .243, p < .05 and β .249, p < .05, respectively); implying that interns’ desire to remain with the company because they feel a sense of attachment and belonging to the organization (AOC), combined with their time invested in the company (COC) plays a significant role in their decision to ultimately accept a job offer for full-time employment from their internship company upon graduation (see Table 4 - Appendix E for a summary of the hypothesized relationships results).

Discussions and Conclusions

The results of this study have numerous implications for developing and implementing internship experiences. First, results indicate the importance for interns to feel a sense of belonging or emotional attachment (i.e., affective organizational commitment) to the company. This is demonstrated by the significant relationships between affective organizational commitment and the following: psychological contract outcomes of employer obligations, job satisfaction, and supervisory support. Our findings are consistent with those of Gault et al. (2000) who found that interns, who felt a strong sense of attachment and belonging to their internship employer, also appeared to be the most satisfied with their internship experience. Accordingly, it is evident that one of the most important sources of interns’ satisfaction revolves around their sense of belonging, it is imperative for supervisors and staff to integrate interns into the operations of the particular location/office to which they are assigned and entrust them with interesting and meaningful tasks. According to Dixon et al. (2005), being assigned challenging work and meaningful tasks is interpreted as an indication of the supervisor’s respect for and trust in the intern’s abilities. Therefore, it is plausible that such respect and trust enhances the overall internship experience

The positive impact that affective and continuance organizational commitment have on interns’ conversion intentions (see Figure 1) implies that interns’ sense of attachment/belonging, combined with their time invested in the company plays a significant role in their decision to accept an offer for full-time employment with the intern company upon graduation. When newcomers (i.e., interns) successfully adjust to their roles and work environment, they are more likely to have a strong attachment (i.e., commitment) to the company, and should be less likely to quit (Wanous, 1992).

Therefore, if companies want to enhance intern’s level of commitment and increase conversion intentions, they need to implement programs that assist interns’ with both task and social transitions. Companies can structure the internship program in a way that introduces new challenges (i.e., new tasks/job duties or skill sets) and responsibilities (i.e., work deadlines, management) each week. By providing the opportunity to learn new skills and have increased responsibilities each week, interns can properly acclimate and socialize into their work role without being overwhelmed; thus, creating a more enriched and meaningful internship experience.

Results of our study reveal that the same factors (i.e., job satisfaction, organizational commitment and supervisory support) that tend to reduce turnover for regular employees have similar influences on recruiting interns (i.e., conversion intention ns). However, if firms use internships as a recruitment tool, it is important that they consider the proper allocation of resources (i.e., mentoring, training and development) in managing the internship experience, so that only the best interns are selected and cultivated for career advancement with the firm. Ultimately, findings of the current study are particularly relevant to companies as they develop and structure internship programs in an attempt to enhance their conversion rates. Some strategies that could provide augmentation of internship conversion and maximize the possibility of a successful internship program include: 1) Treat interns as a part of the organizational team and invite them to staff meetings, 2) Involve interns in project planning and ask for their ideas or suggestions, 3) Hold interns accountable for projects and deadlines, 4) Assign projects that are challenging, yet accomplishable, 5) Assign a supervisor or mentor to each intern to provide the necessary guidance, training, and feedback, and 6) Establish a process for permanent hire considerations and share that information with interns (Coco, 2000). These are actions within a supervisor’s control that can substantially increase the possibility of a successful internship experience benefiting both the intern and the company (Pecorella & Stonecash, 2007).

Limitations and Further Research Suggestions

The sampling frame and on-line survey presented a couple limitations: 1) We cannot control the number of respondents nor the time frame in which the participants respond, and 2) It is difficult to calculate an accurate response rate since we do not have an exact count of how many interns received the survey initially. Also, as with most survey research, a common method bias may be present since all of the data were collected using self-report measures. Additionally, a shortcoming of the current study is the fact that we do not have feedback or performance measures from employers; therefore, creating a direction for future research.

As well, in a fully recursive model, the following direct paths were also tested and revealed nonsignificant path coefficients: SS → COC (.105, t-value = 1.71), SS → CI_A (-.028, t-value = -.61) JS_job → CI_A (.072, t-value = .71) and poor model fit (x ² = 7.004, df = 2, p = .030; RMSEA = .100). Therefore, future research will replicate the study with another sample in an effort to continually refine item measures that capture the internship experience.

Moreover, given the tremendous costs associated with turnover (i.e., lost productivity, hiring and training costs), understanding the causes of turnover are extremely important for organizations (DeConinck & Bachmann, 2005). However, it should be noted that not all turnover is dysfunctional (losing valued employees); functional turnover rids the company of employees that are not performing well and employees who do not fit within the organization (DeConinck & Bachmann, 2005). Therefore, using the factors that contribute to functional and dysfunctional turnover as a guide, future research should examine if the same factors contribute to functional (interns who do not receive offers) versus dysfunctional (interns who receive an offer but do not accept) conversion intentions of interns. Understanding the underlying factors of conversion intentions and interns’ reasons for accepting or declining an offer would help firms minimize their chances of having dysfunctional turnover. Furthermore, if the organization is using internships to recruit, perhaps using the same factors that have been shown to reduce turnover would be applicable to the investigation of the factors that contribute to interns’ intent to accept a job offer after an internship.

Future research will also continue to collect internship data from a larger sample. Since one of the primary functions of the internship is to provide reality testing during college, future research should examine what aspects of the internship provide the most realistic expectations (i.e., work deadlines, dealing with problems on the job, interaction with co-workers, etc.).