Tellers versus technology in overall consumer satisfaction with banking services.

Simmers, Christina S. ; Burman, Bidisha ; Haytko, Diana L. 等

INTRODUCTION

Service encounter satisfaction plays an integral role in the determination of overall satisfaction with the firm by the consumer. "Service encounters are critical moments of truth in which customers often develop indelible impressions of a firm. In fact, the encounter frequently is the service from the customer's point of view" (Bitner, Brown and Meuter 2000, p. 139). Shostack (1985) defines a service encounter as "a period of time during which a consumer directly interacts with a service" (p. 243). This interaction can be either with an employee or with the self-service technology of the company.

For interaction with an employee, personalization is an important factor in customer service satisfaction. Personalization involves the politeness, courtesy, and friendliness of an employee's behavior when interacting with a customer and affects customer service satisfaction (Mittal and Lassar 1996). However, self-service technology has created dramatic changes in the way a company interacts with its customers. According to Bitner, Brown and Meuter (2000), the number of encounters a customer has with a firm may dramatically increase due to the use of technology. The impact of such changes on consumers' perception and level of satisfaction with the service provided by the firm is of interest. In this research, we examine the relative impact of the human encounter and the technological encounter on consumers' overall satisfaction.

THEORETICAL BACKGROUND

Overall Satisfaction Versus Encounter-Specific Satisfaction

Bitner and Hubbert (1994) define service encounter satisfaction as the consumer's dis/satisfaction with a discrete service encounter (e.g., a haircut, an interaction with a dentist, a discussion with a repair person, an experience at a hotel check-in desk) while overall satisfaction is an accumulation of all previous experiences with a service provider, and therefore is a function of all previous encounter-specific satisfaction (Jones and Suh 2000; Bitner and Hubbert 1994). Bitner and Hubbert (1994) found that encounter-specific satisfaction and overall satisfaction are distinct to consumers, but highly correlated. Empirical support was found for this distinction by Jones and Suh (2000), even when measuring both types of satisfaction with the same scale. Bitner et al. (2000) also suggested that the "service from a customer's perspective may actually be a relationship made up of repeated, similar service encounters" and that "each individual encounter can be critical in determining the customer's future behavior toward the company" (p. 139). Since, these encounters may be human encounters as well as technological encounters; we intend to examine the weight of human encounter satisfaction versus technological encounter satisfaction in determining consumers' overall satisfaction level with the firm.

Encounter specific satisfaction and its influence on overall satisfaction can be explained using attribution theory. According to Folkes (1984), attribution theory views people as rational information processors whose actions are influenced by their causal inferences. Prior research demonstrates that only in the case of a service failure do consumers look into the causes of that failure, however the cause is not considered in a successful transaction (Weiner 2000). "Attributions are what people perceive to be the causes behind their own behavior, the behaviors of others, or the events they observe" (Bitner 1990, p. 70). According to Bitner (1990), "Weiner's long stream of research on attributions has led to the conclusion that most cases can be classified on three dimensions: locus (Who is responsible?), control (Did the responsible party have control over the cause?), and stability (Is the cause likely to recur?)" (p. 70). Folkes (1984) states that this classification has been linked to behavioral consequences.

Human Encounters Versus Technological Encounters

In 1994, Kotler developed a model to represent services marketing (Kotler 1994). The Triangle Model of services marketing consists of three key points: the company, employees and customers (Kotler 1994). In 1996, in order to reflect the growing importance of technology in the services industry, Parasuraman extended Kotler's Triangle Model by adding a fourth dimension, technology, to form a Pyramid Model (Parasuraman and Grewal 2000). The base of the pyramid reflects the dynamic relationships among employees, customers, and technology (Bitner, Brown and Meuter 2000). The Pyramid Model emphasizes the critical linkages between each of these parties in determining the success of the service organization.

The satisfaction the customer derives from an interaction with an employee of the company, referred to as human encounter satisfaction, plays an important role in consumers' overall satisfaction with the services of the company. Social exchange theory implies that perceived justice (i.e., the consumers' perception of being treated fairly) affects customer satisfaction. Carr (2007) has incorporated justice theory as the optic from which to view the relationship between the consumer and the service provider and the resulting service satisfaction. His findings using his FAIRSERV model add significant, new predictors of service satisfaction than those previously offered using the traditional SERVQUAL model (Carr 2007). The three dimensions of perceived justice with complaint handling have been identified as distributive, procedural, and interaction justice (Tax et al. 1998). Distributive justice involves dealing with decision outcomes; namely, the principles of equity and equality. Procedural justice involves dealing with decision-making procedures, or having a complaint procedure the customers perceive as fair. Interactional justice entails the interpersonal treatment the customer receives during the enactment of the procedures. Consumer evaluation of the interaction dimension suggests that the quality of the interpersonal treatment and communication during the encounter are likely to be heavily weighted by consumers when evaluating service encounters (Smith, Bolton and Wagner 1999). Human interaction becomes even more important in responding to consumers with special needs, requirements, or preferences where personal attention is required (Bitner, Booms, and Tetreault 1990).

Technological encounter satisfaction, or the satisfaction the customer derives from an interaction with the technology of the company, also plays an important role in consumers' overall satisfaction. The use of technology has permeated all aspects of the services industry (e.g., automated teller machines, pay-at-the-pump gasoline, and grocery self-checkout lines). In these instances, customers actually provide the service for themselves using the self-service technologies of the company, without any employee involvement (e.g., automated teller machines, E*Trade, or online ticketing) (Bitner, Brown and Meuter 2000). Self-service options can either completely replace or complement the services of the organization (Bitner, Brown and Meuter 2000). The main concern of service organizations utilizing more technology is that a technological failure is highly apparent because the transaction cannot be completed (i.e., money could not be drawn from the bank). However, human failure can be considered more ambiguous because, while the service does not meet expectations, the transaction can still be completed (Meuter, 2000).

Consistent with the implications of attribution theory, we propose that human encounters will play a larger role in predicting overall satisfaction than technological encounters. In a technological (self-service) transaction, consumers will accept responsibility themselves and, due to the lack of a third party to blame, they are less likely to be upset by a service failure. Overall, we posit that consumers are likely to accept more responsibility for the outcome of self-service technology failure and, hence, may continue usage of technology as a service provider in the future. On the other hand, they will be more dissatisfied with a failure in a transaction involving a human encounter in which they can blame the employee. Similarly, a successful transaction via a human encounter will carry more weight as compared to a successful transaction via technology. Therefore, it is posited that, although each type of encounter satisfaction is predicted to be significant, the human encounter will carry more weight in consumer evaluation of overall service satisfaction.

Hypothesis 1: Satisfaction with the human encounter will have a stronger impact on overall satisfaction than satisfaction with the technological encounter

STUDY ONE METHOD

Dependent Variable

The dependent variable of interest is overall satisfaction with bank services. In order to operationalize overall satisfaction, respondents were asked to think about all prior experiences with that specific bank. Six items were used to measure overall satisfaction using five-point Likert scales. One item asked respondents to indicate how satisfied or dissatisfied they were with the service at their bank based on all their experiences. The remaining five items were statements to which the respondents indicated their level of agreement or disagreement.

Independent Variables

The independent variables identified included human encounter satisfaction and technological encounter satisfaction. Human encounter satisfaction was operationalized as the last encounter with a bank teller. This was measured with six items using five-point Likert scales. Respondents were presented with statements to which they indicated their level of agreement or disagreement to their encounter with a bank teller. Technological encounter satisfaction was operationalized as the last encounter with a bank Automated Teller Machine (ATM), and was similarly measured with six items using five-point Likert scales. Again, respondents were presented with statements to which they indicated their level of agreement or disagreement to their encounter with an ATM.

Subjects and Procedure

The surveys were administered to undergraduate students from six business classes for extra credit. Banking was selected to be the context to be used to test the hypothesis. Data were collected from 194 undergraduate university students who have a bank checking account. Students were told that a survey was being conducted on bank services. Consent forms were collected and confidentiality was also ensured. Students were

advised to read the instructions carefully and to take the time necessary to fill out the survey completely. The measurement instrument was adapted from the work of Bitner and Hubbert (1994) with a few minor modifications.

The first section asked questions regarding the name of the bank in which the respondent had a checking account, the length of time the respondent had this account, and the frequency in which the respondent used the teller and ATM services of that bank. Sections 2 and 3 each consisted of six questions related to satisfaction with the last encounter the respondent had with a bank teller and a bank ATM machine, respectively. Section 4 consisted of overall satisfaction with the bank, taking into consideration all experience with that bank. Section 5 consisted of questions related to computer and Internet usage of the respondent and were not examined in this study. Section 6 collected demographic information of the respondents. The instrument was pre-tested and minor modifications were made.

RESULTS

The descriptive statistics were examined to establish the demographic profile of the respondents. Most of the respondents are between the ages of 19 and 21 years and are classified as juniors at the university. Each gender is equally represented in the study.

Data was analyzed to determine banking behavior of the respondents. They indicated having a checking account at one of 31 different banks. Results indicate that 74.7 percent of the respondents have held their checking account at their present bank between one and five years, 14.9 percent for less than one year, and 9.8 percent for six to ten years. Also, 80.9 percent of the respondents have used the teller service at their bank and 86.1 percent have used the ATM.

The reliability tests showed high Cronbach's alphas for overall satisfaction ([alpha].91), human encounter satisfaction ([alpha].90), and technological encounter satisfaction ([alpha].90). Multiple regression analysis was conducted using the stepwise approach to determine the influence of human and technological encounter satisfaction on overall satisfaction. The multiple regression results are shown in Tables 1(a) and 1(b). The results in Table 1(a) show that the overall model is significant (F=13.418, p<.001), indicating that both types of encounter satisfaction predict overall satisfaction. However, in support of the hypothesis, the beta coefficients indicate that human encounter satisfaction (t = 4.475, p< .001,[beta]= .331) is a much stronger predictor of overall satisfaction than technological encounter satisfaction (t = 1.987, p = .049, [beta] = .147).

As displayed in Table 1(b), both human encounter satisfaction and technological encounter satisfaction are associated with overall satisfaction explaining 14 percent of the variance in the dependent variable (R square = .144). The increase in the R square (.123 to .144) shows that both human and technological encounter satisfaction explain more variance in the dependent variable than human encounter satisfaction alone. While the incremental contribution of technological encounter satisfaction is significant (p=.049), the R square change (.021) is small.

Additional Analysis

Additional analysis was conducted by dividing the respondents into two groups according to their preference for service provided by a human being and self-service technology. Regression was run for each group separately. The results shown in Tables 2(a) and 2(b) indicate that for those respondents that prefer human interaction, satisfaction with technology has no effect (t=-.188, p=.851,[beta]=- .021) and human encounter satisfaction is the only predictor of overall satisfaction (t=3.597, p=.001,[beta]= .402). The [R.sup.2] change remains the same ([R.sup.2] change = .000), indicating that technology encounter satisfaction explains no more variance in the dependent variable than human encounter satisfaction ([R.sup.2] change = .156). This is likely attributable to the fact that these individuals hardly use self-service technology.

However, the respondents who prefer technological interaction revealed interesting results (see Tables 3a and 3b). While technological encounter satisfaction was significant (F=7.112, p=.001), results showed that for these respondents human interaction was also considered more important ([beta] = .329) than technological encounter satisfaction ([beta] = .259). Further, a greater [R.sup.2] change was found for technological encounter satisfaction ([R.sup.2] change = .066) than compared to the overall model ([R.sup.2] change = .021) (see Table 1b).

DISCUSSION

The purpose of this study was to establish the importance of human encounter satisfaction as compared to technological encounter satisfaction in overall service satisfaction. Although technology may increase the number of encounters the customer has with the firm (Bitner, Brown and Meuter 2000) and enhance customer satisfaction with every successful encounter, the human element is not to be ignored. The additional analysis further revealed that for individuals who prefer human encounters, technological encounter satisfaction does not play a role in determining their overall satisfaction with the service. However, for individuals who prefer technological encounters, human encounter satisfaction is still more important in determining their overall satisfaction. These findings provide further support for the hypothesis.

Considering the increase in the use of online banking services ("What" 2001), a second study was conducted to include this popular form of technological interaction. In addition to the independent variables in Study One, banking service encounter satisfaction to overall service satisfaction will be examined.

STUDY TWO METHOD

Dependent Variable

Again, overall satisfaction with bank services is the dependent variable. Study Two utilized the same six-item, five-point Likert scales used in Study One.

Independent Variables

The independent variables was again the human encounter satisfaction and technological encounter satisfaction. Human encounter satisfaction was operationalized using the same measures used in Study One. However, two types of technological encounter satisfaction were measured in Study Two. Using the same six items and five-point Likert scales as in Study One, the last encounter with the bank ATM and the last encounter with the online banking service operationalized technological encounter satisfaction.

Subjects and Procedure

Surveys were administered to one undergraduate and two graduate business classes. Banking was again used as the context. Data were collected from 153 respondents who have a bank checking account. The same procedures were used as in study one.

RESULTS

Descriptive statistics indicated the respondents ranged from 19 to 61 years of age with a mean of 24. Fifty-two percent of the respondents were female.

Data was examined to profile the banking and behavior of the respondents. Respondents indicated having a checking account at one of 35 different banks. Results indicate that 55.3 percent of the respondents have held their checking account at their present bank between one and five years, 23 percent between six and ten years, 15.1 percent for less than one year, and 6.6 percent for greater than 10 years. Also, 80.4 percent of the respondents have used the teller service at their bank, 85.8 percent have used the ATM, and 72.4 percent have used the online banking service. With the addition of online banking, the online behavior of the respondents was examined. Respondents frequently use the online banking service (mean = 4.17), checking their account balance online (95 percent) and paying their bills online (57.5 percent). The respondents surf the internet several times daily (72.4 percent), buy products on the internet (96.1 percent), are comfortable using the computer (mean = 4.72) and use the computer frequently in their daily life (mean = 4.73).

The reliability tests showed high Cronbach's alphas for the variables of interest, including teller encounter satisfaction ([alpha].87), ATM encounter satisfaction ([alpha].94), online encounter satisfaction ([alpha].91), and overall satisfaction ([alpha].91). Multiple regression analysis was conducted using the stepwise approach to determine the influence of teller encounter satisfaction, ATM encounter satisfaction, and online encounter satisfaction on overall satisfaction. The multiple regression results are shown in Tables 4(a) and 4(b). The results in Table 4(a) show that the overall model is significant (F = 21.761, p < .000), indicating that teller encounter satisfaction and online encounter satisfaction predict overall satisfaction. ATM encounter satisfaction was not entered into the model. Study Two results do not support the hypothesis that human encounter satisfaction is a stronger predictor of overall satisfaction than technological encounter satisfaction. The beta coefficients indicate that online encounter satisfaction (t = 4.74, p < .000,[beta]= .425) has greater weight than teller encounter satisfaction (t = 3.05, p < .003,[beta]= .274) on overall satisfaction, although both play a role in overall satisfaction.

As displayed in Table 4(b), both human encounter satisfaction (teller encounter satisfaction) and technological encounter satisfaction are associated with overall satisfaction, explaining 32.1 percent of the variance in overall satisfaction (R square = .321). The increase in the R square (.252 to .321) shows that both human and technological encounter satisfaction explains more variance in overall satisfaction than online satisfaction alone.

To further analyze the lack of contribution of ATM encounter satisfaction to overall satisfaction, another multiple regression analysis was conducted using the enter approach which forces all three proposed predictors into the regression model. The multiple regression results (see Table 5a) show that the overall model is significant (F = 14.471, p < .000), including the predicting variables of teller satisfaction, ATM satisfaction, and online satisfaction. Again, the beta coefficients indicate that online satisfaction (t = 4.53, p < .000,[beta]= .416) and teller satisfaction (t = 3.01, p < .003,[beta]= .271) are predictors of overall satisfaction. ATM satisfaction is non-significant (t = 0.50, p = .619,[beta]= 0.044).

As displayed in Table 5(b), teller satisfaction, ATM satisfaction, and online satisfaction together explain 32.3 percent of the variance in overall satisfaction (R square = .323). Appropriately, the total variance explained is relatively the same whether using the stepwise approach (32.1 percent) or the enter approach (32.3 percent).

ADDITIONAL ANALYSIS

Respondents were again grouped according to their preference of either self-service technology or service provided by a human being. Regression using the stepwise approach was run for each group. As in the previous regressions, ATM encounter satisfaction was not entered into the models. As a result, for the remainder of the paper, human encounter satisfaction will be represented by teller encounter satisfaction and technological encounter satisfaction will be represented by online banking service encounter satisfaction.

For those who prefer technological interaction, both human and technological encounter satisfaction impact overall satisfaction (see Tables 6a and 6b). Technological encounter satisfaction is more important (t = 3.31, p = .002,[beta]= .401) followed closely by human encounter satisfaction (t = 3.00, p = .004,[beta]= .363) (see Table 6a). As shown in Table 6(b), the addition of human encounter satisfaction increases the R square (from R square = .296 to R square = .408) and the R square change is significant (R square change = .111, F = .004).

Oddly, for those who prefer human interaction (service provided by a human being), human encounter satisfaction has no effect by not even entering the model (F = 11.044, p = .002). Technological encounter satisfaction is the only predictor (t = 3.32, p = .002,[beta]= .461). Table 7a shows these results. Table 7(b) shows technological satisfaction alone explains 21.2 percent of the variance in overall satisfaction (R square = .212) for those who prefer human interaction.

DISCUSSION

The purpose of the second study was to examine the impact of online encounter satisfaction on the relationship examined in Study One. Study Two demonstrates the impact of the online banking service encounter on overall service satisfaction. In this case, technological encounter satisfaction has greater impact on overall service satisfaction than human encounter satisfaction, counter to the findings in the first study. However, human interaction remains an important determinant in overall service satisfaction.

IMPLICATIONS

The purpose of the two studies was to examine the relative importance of human encounter satisfaction and technological encounter satisfaction on overall service satisfaction. Although technology may increase the number of encounters the customer has with the firm (Bitner et al. 2000), the human element is not to be ignored. With the prevalence of online communication and transactions, the pyramid model (Parasuraman and Grewal 2004 provides support for the importance of the dynamic relationship between employees, customers and technology (Bitner, Brown and Meuter 2000). The human encounter still plays a role but even for those who prefer it, the technological encounter is becoming even more important. Rising dependence on these technologies and the resulting increase in the number of encounters increases the importance of satisfaction with technological interactions on overall satisfaction with the firm.

The managerial implications of these findings indicate that service organizations should continue to pay attention to training their employees who have direct contact with their customers in order to improve their customers' overall satisfaction. However, satisfaction with technological communication is an indispensable and a significant predictor of overall satisfaction. Bitner, Brown and Meuter (2000) assert that offering both types of service encounters, technologically or interpersonally based, is critical to ensure the overall satisfaction of customers. Further, customers are contacting these service organizations through multiple channels so a plan for integrating service across channels is necessary (M.D.F 2007).

There are some limitations of this study that may have influenced the results. First, there is a concentration of ages due to the samples selected. Second, the study is geographically specific. Third, the study focused only on retail banking transactions. Future research may take other regions, ages and contexts into consideration. Further, usage was not measured in this study and may be examined in future research. The research would also benefit from examining the different predictors of satisfaction are for each of the channels customers utilize (van Birgelen et al. 2006; Zhang et al. 2006).

In sum, technological advances have allowed service organizations to offer its customers the convenience of self-service technology in addition to the personalized service of an employee. "Just as continual training and investment in front-line employees helps improve service delivery, self-service technologies must receive on-going maintenance to ensure continued effectiveness (Bitner, Brown and Meuter 2000, p. 59)." By enabling customers to select the type of service encounter they prefer, technological or interpersonal, they can experience the service as they desire, improving overall service satisfaction.

REFERENCES

Bitner MJ. (1990), Evaluating Service Encounters: The Effects of Physical Surroundings and Employee Responses, Journal of Marketing, 54(April): 69-82.

Bitner MJ, BH Booms & MS Tetreault (1990), The Service Encounter: Diagnosing Favorable and Unfavorable Incidents, Journal of Marketing, 54(January): 71-84.

Bitner MJ & AR Hubbert (1994), Encounter Satisfaction Versus Overall Satisfaction Versus Quality. In: Rust, RT, Oliver, RL, (Eds.). Service Quality: New Directions in Theory and Practice. Thousand Oaks, CA: Sage, pp. 72-94.

Bitner MJ, SW Brown & ML Meuter (2000), Technology Infusion in Service Encounters, Journal of the Academy of Marketing Science, 28(1): 138-149.

Carr, CL. (2007), The FAIRSERV Model: Consumer Reactions to Services Based on a Multidimensional Evaluation of Service Fairness, Decision Sciences 38(1): 107-130.

Folkes VS. (1984), Consumer Reactions to Product Failure: An Attributional Approach, Journal of Consumer Research, 10(March): 398-409.

Jones MA & J Suh (2000), Transaction-Specific Satisfaction and Overall Satisfaction: An Empirical Analysis, Journal of Services Marketing, 14(2): 147-159.

Kotler, P. (1994), Marketing Management: Analysis, Planning, Implementation, and Control. 8th ed. Englewood Cliffs, NJ: Prentice Hall.

Meuter ML, AL Ostrom. RI Roundtree & MJ Bitner (2000), Self-Service Technologies: Understanding Customer Satisfaction with Technology-Based Service Encounters, Journal of Marketing, 64(July): 50-64.

M.D.F. (2007), Multichannel Service Improves ... Or Does It?, Multichannel Merchant, 3(Mar): 54.

Mittal B & WM Lassar (1996), The Role of Personalization in Service Encounters, Journal of Retailing, 72(1): 95-110.

Parasuraman A & D. Grewal (2000), The Impact of Technology on the Quality-Value-Loyalty Chain: A Research Agenda, Journal of the Academy of Marketing Science, 28(1): 168-174.

What We Have Here Is Channel Evolution, (2001), ABA Banking Journal, 93(9): 514.

Shostack GL. (1985), Planning the Service Encounter. In: Czepiel JA, Solomon MR, Suprenant CF, (Eds.). The Service Encounter. New York: Lexington Books, pp. 243-254.

Smith AK, RN Bolton & J. Wagner (1999), A Model of Customer Satisfaction With Service Encounters Involving Failure and Recovery, Journal of Marketing Research, 36(3): 356-363.

Tax, SS, SW Brown & M. Chandrashekaran (1998), Customer Evaluations of Service Complaint Experiences: Implications for Relationship Marketing, Journal of Marketing, 62(April): 60-76.

Weiner B. (2000), Reflections and Reviews: Attributional Thoughts About Consumer Behavior, Journal of Consumer Research, 27(December): 382-387.

Van Birgelen, M. (2006), Multi-Channel Service Retailing: The Effects of Channel Performance Satisfaction on Behavioral Intentions, Journal of Retailing, 82(4): 367-377.

What We Have Here Is Channel Evolution, (2001), ABA Banking Journal, 93(9): 514.

Zhang X, V. Prybutok & A.Huang (2006), An Empirical Study of Factors Affecting E-Service Satisfaction, Human Systems Management, 25(4): 279-291.

Christina S. Simmers, Missouri State University Bidisha Burman, Appalachian State University Diana L. Haytko, Missouri State University Christopher A. Ellis, Missouri State University

Table 1(a): Multiple Regression Results

Model F Sig. Beta t Sig.

(a) Overall Model 13.418 .000
 Human Satisfaction .331 4.475 .000
 Satisfaction Technological .147 1.987 .049

(a) Predictor: Human Encounter Satisfaction, Technological
Encounter Satisfaction Dependent Variable: Overall
Satisfaction

Table 1(b): Multiple Regression Results

Model R R Square Adjusted R Square Change Significant
 R Square F Change

1 (a) 0.35 0.12 0.12 0.12 0.00
2 (b) 0.38 0.14 0.13 0.02 0.05

(a) Predictor: Human Encounter Satisfaction

(b) Predictor: Human Encounter Satisfaction, Technological
Encounter Satisfaction Dependent Variable: Overall
Satisfaction

Table 2(a): Multiple Regression Results

Human Interaction Preference Group

Model F Sig. Beta t Sig.

(a) Overall Model 7.296 .001
 Human Satisfaction .402 3.597 .001
 Technological -.021 -.188 .851
 Satisfaction

(a) Predictor: Human Encounter Satisfaction, Technological
Encounter Satisfaction Dependent Variable: Overall
Satisfaction

Table 2(b): Multiple Regression Results

Human Interaction Preference Group

Model R R Square Adjusted R Square Significant
 R Square Change F Change

1 (a) .394 .156 .145 .156 .000
2 (b) .395 .156 .135 .000 .851

(a) Predictor: Human Encounter Satisfaction

(b) Predictor: Human Encounter Satisfaction,
Technological Encounter Satisfaction Dependent
Variable: Overall Satisfaction

Table 3(a): Multiple Regression Results

Technological Interaction Preference Group

Model F Sig. Beta t Sig.

(a) Overall Model 7.112 .001
 Human Satisfaction .329 3.113 .003
 Satisfaction Technological .259 .259 .017

Predictor: Human Encounter Satisfaction, Technological
Encounter Satisfaction Dependent Variable: Overall
Satisfaction

Table 3(b): Multiple Regression Results

Technological Interaction Preference Group

Model R R Square Adjusted R Square Significant
 R Square Change F Change

1 (a) .303 .092 .080 .092 .007
2 (b) .397 .158 .136 .066 .017

(a) Predictor: Human Encounter Satisfaction

(b) Predictor: Human Encounter Satisfaction,
Technological Encounter Satisfaction

Table 4(a): Multiple Regression Results
Stepwise Approach

Model F Sig. Beta t Sig.

(a) Overall Model 21.761 .000
 Online Satisfaction .425 4.740 .000
 Teller Satisfaction .274 3.050 .003

(a) Predictor: Online Encounter Satisfaction, Teller Encounter
Satisfaction Dependent Variable: Overall Satisfaction

Table 4(b): Multiple Regression Results
Stepwise Approach

Model R R Square Adjusted R Square Significant
 R Square Change F Change

1 (a) .502 .252 .244 .252 .000
2 (b) .567 .321 .306 .069 .003

(a) Predictor: Online Encounter Satisfaction

(b) Predictor: Online Encounter Satisfaction, Teller Encounter
Satisfaction Dependent Variable: Overall Satisfaction

Table 5(a): Multiple Regression Results
Enter Approach

Model F Sig. Beta t Sig.

a Overall Model 14.471 0.000
Teller Satisfaction 0.271 3.010 0.003
ATM Satisfaction 0.044 0.50 0.619
Online Satisfaction 0.416 4.530 0.000

(a) Predictor: Teller Encounter Satisfaction, ATM Encounter
Satisfaction, Online Encounter Satisfaction Dependent
Variable: Overall Satisfaction

Table 5(b): Multiple Regression Results
Enter Approach

Model R R Square Adjusted R Square Significant
 R Square Change F Change

1a .568 .323 .301 .323 .000

(a) Predictor: Teller Encounter Satisfaction, ATM Encounter
Satisfaction, Online Encounter Satisfaction Dependent
Variable: Overall Satisfaction

Table 6(a): Multiple Regression Results
Technological Interaction Preference Group

Model F Sig. Beta t Sig.

(a) Overall Model 16.509 .000
 Technological Satisfaction .401 3.310 .002
 Human Satisfaction .363 3.000 .004

(a) Predictor: Technological Encounter Satisfaction, Human Encounter
Satisfaction Dependent Variable: Overall Satisfaction

Table 6(b): Multiple Regression Results
Technological Interaction Preference Group

Model R R Square Adjusted R Square Significant
 R Square Change F Change

1 (a) .544 .296 .282 .296 .000
2 (b) .638 .408 .383 .111 .004

(a) Predictor: Technological Encounter Satisfaction

(b) Predictor: Technological Encounter Satisfaction, Human
Encounter Satisfaction Dependent Variable: Overall Satisfaction

Table 7(a): Multiple Regression Results
Human Interaction Preference Group

Model F Sig. Beta t Sig.

(a) Overall Model 11.044 .002 3.320 .002
 Technological Satisfaction .461

(a) Predictor: Technological Encounter Satisfaction
Dependent Variable: Overall Satisfaction

Table 7(b): Multiple Regression Results
Human Interaction Preference Group

Model R R Square Adjusted R Square Significant
 R Square Change F Change

1 (a) .461 .212 .193 .212 .002

(a) Predictor: Technological Encounter Satisfaction
Dependent Variable: Overall Satisfaction