INTRODUCTION
Structural Equation Modeling (SEM) is a comprehensive statistical method to test hypotheses about causal relationships between the observed and unobserved (latent used) variables and has proven itself in solving the problems in the formulation of theoretical constructs (Reisinger, 1999) . Its function is found to be better than other multivariate statistical techniques, including multiple regression, path analysis and factor analysis. Other statistics techniques could not be taken into account, which hang in the interaction between the effects and independent variables. Therefore, a method that can examine a number of dependency relations is also helping to address complex management and behavioral issues. SEM also expand the reasoning ability and statistical efficiency for model with a single, comprehensive method (Pang, 1996).
Baumgartner and Steenkamp (2000) on the role of SEM in marketing modeling and management decisions. Discuss them some advantages. They said that although SEM has potential to modeling decisions, it is probably most useful for the theoretical test, which is an important phase in the development of marketing models [For SEM and LISREL see, Byrne (1998), Cheng ( 2001), et al Cudeck. (200), Hayduk (1987), Joreskog and Sorbom (2001)].
Applied to data on attitudes, perceptions, said the behavior intentions and actual behavior, SEM can be used to specify and test alternative causal hypotheses. It has been found that, predictably, causality is often mutually exclusive. The assumption that the behavior of attitudes, perceptions and behavioral intentions influence without feedback does not hold up when tested SEM. These results represent the assumption held by some that we found that this setting decisions can be determined directly scaled in preference choice models. It was used path analysis of empirical evidence that the causal link of behaviors, attitudes to choice is more show than the combination of settings to choose behavior. Subsequent studies with different forms of simultaneous equation modeling showed consistently that attitudes, particularly the perception of limited choices, while at the same time, attitudes, decisions (Golob affect 2001b) to. Gärling et al. (2001) examines the decision-making with decisions of the trip with a SEM with latent variables, the links between attitude towards driving, frequency of choice of the driving test and revealed the existence of a certain type of decision-making process known as a script-based. Golob (2001a) tested a series of joint models of attitude and behavior to explain how both mode choice and attitude. Stopped applying weighted least squares (WLS) estimation on a data from San Diego, California, the author shows that seem to influence decisions that some opinions and perceptions, but also different opinions and perceptions of the behavior and dependent only on exogenous variables, personal and household goods. None of the models tested found any significant impact on the choice of settings.
Most newspapers have written on the variables to explain the attractiveness of the shopping choice include [For exammple, Suarez et al. (2004), Degeratu et al. (2000), Severin et al. (2001)]. They have always logit models and random effect model. Degeratu et al. (2000) specifically to evaluate whether brand name and price have an impact on choices online and traditional supermarkets to concentrate. Severin et al. (2000) investigated the use of relatively new developments in random utility theory to assess the stability over time and space, the preferences of retail shopping choices. They found that a good quality, great selection, good service, were pleasant atmosphere and convenient location significant choice of retail shopping center model. They noted that high and low prices and the latest fashion not always significant in each year models. It also showed that the optimal position had the greatest impact on the shopping decisions.
METHOD
The study was considered the research factors that consumers in choosing shopping centers and developed a proposal to develop a model for shopping center choice. In addition to demographic questions, effective factors that have been made to Cart peoples’ choice and center for 17 products have been) very important, with answers taken from five-point Likert scale (5 = very important and. These points are given in Table 1.
Reliability coefficient of the questionnaire was calculated as Cronbach = 0. 79. If the components of reduced alpha reliability coefficient value were deleted increased to 0. 81. In this study, latent structure is in the choice of shopping center (E) and explanatory structures of features of materials sold (A), consisting of attitude and behavior of staff (B) Location of Shopping Centers (C), alleviation of Price (D), the regularity at the shopping center (F). The structure of the relationship of the judge accepted 5 independent latent variables (A, B, C, D and F) on a latent variable (E) are to be tested by the model. Hypotheses developed to test the relationship between the latent constructs are given below:
H1; There is a significant correlation between the choice of shopping center and the properties of the materials sold in the middle.
H2; There is a significant correlation between the choice of shopping and attitude and behavior of employees,
H3; There is a significant correlation between the choice of shopping centers and the geographic location of the center.
H4; There is a significant correlation between the choice of shopping center and facilitate the price.
H5; There is a significant correlation between the choice of shopping and regularity at the shopping center.
RESULTS AND DISCUSSION
Adopted in this study, four models for latent variables associated with the choice of a shopping center were tested with LISREL computer program concerned with SEM. Initially, model M1, was held in which all independent variables are analyzed. Analysis results are given in Table 1. When Table 1 Analysis results are examined for M1, we see that A, B, D and F are not significant latent variables, goodness of fit index to acceptable levels and explanatory power is 52% of the area. M1 path diagram is shown in Figure 1. Finally, if the results of M2, which are found by subtracting observed B, D, a model of F, it is seen that A and C parameter estimates are of significance and fitness criteria are to be kept in acceptable limits. R2 values of the tested models are calculated as 0. 52 and 0 77 and. When the best right model, M2 is observed, 77% of the latent variable, the shopping center is the choice is explained by A and C independent latent constructs. H2, H4 and H5 assumptions for M2 was not approved. Path diagram for M2 is shown in Figure 2, the model parameter estimates and t values are given in Table 2. Parameter estimates for AE and CE relationships in Table 2 and Figure. 2 0 are. 67 and 0 50, respectively. These coefficients are positive and statistically significant. Models analyzed results show that close proximity to the address, discount card application, market presence and facilitating access to the shopping centers, or are the priority in preference of consumer choice of shopping center. In addition to advice advertising for the mall and neighbors important role in the election. The results showed that the behavior of sales personnel at the mall and increase the amount to be preferred cards, on the other side, easement of access and proximity to their addresses as a priority in the decision for shopping centers, compared to servitude of the price.
CONCLUSION
Finally, latent variable, the choice of shopping centers, may be explained at a rate of 77% of independent latent variables, ie properties of materials sold and the geographic location of the shopping center. The importance of the unexplained part of 23% is that the consumer center in relation to other factors that are not taken into account in this study to choose shopping cart. The M2 was the best model in the study, is a proposed model, which depends on some set a
mount of data. It is possible to achieve models of high prices by increasing volume of data with alternative models.
Table 1 – item to shopping center of choice available (model 1)
Estimate of the parameter t-value
A-properties of materials sold (A) 0. 341. 42
Brand name of materials sold (A1) 0. 31 3. ** 99
Quality of materials sold (A2) 0. 46 7. *** 80
Low Price (A3) 0. 071. 29
Large selection (A4) 0. 27 5. ** 09
W-recruitment and employee behavior (B) 0. 010. 03
Behavior of sales personnel (B1) 0. 51 11. *** 3
Friendliness of the staff (B2) 0. 53 11. *** 6
C-Location (C) 0. 41 3. ** 08
Close to the address (C1) 0. 67 8. *** 64
Reduction of access (C2) 0. 70 9. *** 45
D-relieving Price (D) 0. 26 * 2. 06
Payment Form (D1) 0. 35 5. ** 83
Promote commercialization (D2) 0. 58 7. ** 30
Discount card (D3) 0. 69 9. *** 22
F-Regularity (F) -0. 06-0. 3
Well-organized (F1) 0. 42 8. *** 45
Moving to the mall
without difficulty (F2) 0. 43 8. *** 36
E-shopping choice (E)
Neighbor advice (E1) 0. 60
Advertising (E2) 0. 59 6. ** 42
Image (E3) 0. 61 6. ** 55
* P ? 0 05, ** p ? 0 01, *** p ? 0 001
Fig. 1. Path diagram for Model M1
Fit: NFI: 0 85, NNFI: 0 87, CFI: 0 90, GFI: 0 91, AGFI: 0 87, ? 2 / df = 2 21
Table 2 – item to shopping center of choice available (model 2)
Estimate of the parameter t-value
A-properties of materials sold (A) 0. 67 ** 4 76
Brand name of materials sold (A1) 0. 61 5. ** 83
Quality of materials sold (A2) 0. 25 ** 4 34
Large selection (A4) 0. 17 * 2. 79
C-Location (C) 0. 50 ** 4 11
Close to the address (C1) 0. 66 7. *** 81
Reduction of access (C2) 0. 71 8. *** 59
E-shopping choice (E)
Neighbor advice (E1) 0. 53
Advertising (E2) 0. 62 6. ** 17
Image (E3) 0. 64 6. ** 29
* P ? 0 05, ** p ? 0 01, *** p ? 0 001
Fig. 2. Path Diagram for Model M2
Fit: NFI: 0 80, NNFI: 0 79, CFI: 0 87, GFI: 0 94, AGFI: 0 88, ? 2 / df = 3 42
REFERENCES
Byrne, B. M. (1998). Structural Equation Modeling with LISREL, Prelis and SIMPLIS: Basic Principles, Applications and Programming, New Jersey: Lawrence Erbaum Associates Publisher.
Cheng, W. E. L. (2001). SEM they are better than multiple regression in Parsimonious Model Testing for Management Devolopment Research. Journal of Management Development. 20 (7), 650-667.
Cudeck, R., Toit, DS & Sörbom, D. (2000). Structural Equation Modeling: Present and Future, Scientific Software International Inc.
Degeratu, AM, Rangaswamy, A. & Wu, J. (2000). The choice of consumers’ online behavior and traditional supermarkets: The effects of the brand, price and other attributes. Intern. J. of Research in Marketing, 17, 55-78.
Hayduk L. A. (1987). Structural Equation Modeling with LISREL Essential and advances. The John Hopkins University Press.
Gärling, T., S. Fujii & Boe, O. (2001). Empirical tests of a model of determinants of script-based driving choice. Transportation Research F 4, 89-102
Golob, T. F., 2001 (a). Joint models of attitudes and behavior in evaluation of San Diego I-15 Congestion Pricing Project. Transportation Research A 35, 495-514.
Golob T. F., 2001 (b). Structural Equation Modeling for Travel Behavior Research. UCI-ITS-AS-WP-01-2, Center for Activity Systems Analysis and Institute of Transportation Studies, University of California, Irvine, Irvine, CA 92697-3600, USA, November 2001, Institute of Transportation Studies, University of California.
Joreskog, K. & Sorbom, D. (2001). LISREL 8: User’s Reference Guide, Scientific Software International Inc.
Pang, S. K. N. (1996). School values and feelings of Teachers: a LISREL model. Journal of Educational Administration. 34 (2), 64-83.
Reisinger, Y. & Turner, L. (1999). Structural Equation Modeling with LISREL: application in tourism. , Tourism Management. . 20, 71-88.
Suarez, A., Rodriguez del Bosque, I., Rodriguez-Poo, JM, & Moral, I., (2004). Accounting for heterogeneity in shopping center choice models. Intern. J. of Research in Marketing, 11, 119-129.
Severin, V., Louviere, J. J & Finn, A. (2001). The stability of retail purchasing decisions over time and between countries. Journal of Retailing, 77, 185-202.
Steenkamp, BEM & Baumgartner, H. (2000). On the use of structural equation models for marketing modeling. Intern. J. of Research in Marketing, 17, 195-202.