Buying decisions carry some level of risk in terms of the uncertain environment faced by the buyer, and the possible negative consequences of the buying process. Even after the online purchase has been made, the buyer will not know whether or not the chosen product or vendor will actually satisfy his/her needs. Buyers will thus attempt to mitigate this perceived risk by assessing different alternatives and selecting the one with the lowest perceived risk [12-14].

It has been found that perceived risk is a major obstacle for buyers considering making an online purchase; an obstacle that is substantially greater than those possibly experienced in physical stores. It is not surprising that consumers will be more reluctant to make online transactions, given the potential losses that may result from the decisions made in uncertain environments [12,14,15].

In traditional stores, buyers can touch, feel, and even try the product before deciding to buy it; this immediately reduces the perceived risk, and probably reinforces positive reviews from consumers about the store. In online purchasing, on the other hand, not only can the buyer not try the product, but s/he also has to provide valuable personal information such as home address, phone number, and even credit card information, and then wait for the transaction to be approved [12-15].

To understand the concept of customer orientation, we first have to understand what Market Orientation (MO) is. MO is the ability of organizations to interact with their environment, allowing them to anticipate market needs and create long-term relationships with their customers, suppliers, and distribution channels [16].

Market orientation makes companies more capable of reaching their objectives; however, for its implementation to be successful, significant cultural transformation is required to develop the values and standards of conduct related to this management approach. MO can be considered a construct consisting of three behavioral components of equal importance: customer orientation, competition orientation, and functional coordination, with two decisional criteria: long-term focus and profit orientation [16].

As such, customer orientation can be defined as a company’s knowledge of its target market in order to be able to create superior value for such a niche. This knowledge enables the organization to explore opportunities for innovation and reduce the potential risks of confusing the buyers’ needs. It calls for a continuous and proactive willingness to meet customer demands [17,18], but could also be used as a tool to influence customers and somehow increase the reputation of the organization. According to the above, it is proposed that the following relations should be verified:

H1: The online vendor´s customer orientation affects social influence

H2: The online vendor´s customer orientation affects the perceived reputation of such vendors.

The concept of social influence is based on the subjective norm proposed in the Technology Adoption Model (TAM), and describes the influence of people who are important to the subject making decisions, even if these other views go against the buyer’s own views in terms of the purchase [19]. This concept has therefore been linked directly with online buying, adding social factors to it [20], [21]. Hence its importance lies in the effect of influential people’s favorable or unfavorable opinions about the web vendor. It can therefore be considered that social influence can directly influence a vendor´s reputation. This study intends to verify the following relationship:

H3: The social influence regarding a web vendor positively influences its perceived reputation.

Reputation is what is generally said or believed about the nature or status of a person or thing. Reputation might somehow mitigate the fears that still surround the Internet as a means of doing business. An organization with a good reputation will beget a more positive attitude towards its e-service offer from its customers, as online buyers will favor sellers already known through traditional channels [14,22-25].

Decisions regarding the reputation of an online provider can be a consequence of accumulated experience with such a provider or they can derive from more emotional bases. As such, a buyer will always prefer an online provider with whom a satisfactory transaction has already been experienced. This suggests that satisfaction in past interactions contributes directly to building reputation and increases each time buyers’ expectations are met or exceeded [14,24,26].

2.4.1. The relationship between web reputation and perceived risk

Previous studies support the direct influence between reputation and purchase intention [27], as well as the perceived value of use [28], and an important predictor of trust [29]. These relations generate direct effects on risk perception because reputation is the indicator that assesses the information the buyer has obtained intrinsically from its interaction with the web vendor and extrinsically by social information that influences and therefore directly affects the perception of risk in online buying [30]. The study also aims to verify that:

H4: Perceived reputation of the web vendor directly influences perceived risk in e-commerce.

Web privacy can be understood as consumers’ concerns about the misuse of the data provided to complete a transaction for other purposes, such as: procurement and inadequate storage of information; distribution to third parties without authorization; invasion of privacy, emailing without prior request, among others. If customers do not perceive that their personal information will be kept private, the purchase will not take place, making this factor an essential part of the decision of whether or not to make an online [13,14,24,31,32].

2.5.1. Relationship between web privacy and perceived risk

Establishing security protocols by businesses and making them visible to consumers can play a major role in reducing perceived risk. Such protocols have to be present both in information gathering and its subsequent use, thereby reducing fear and increasing confidence, so that users are more willing to provide the required personal information [31,33].

Therefore, we propose to verify the following relationship:

H5: Web privacy negatively influences perceived risk.

Finally, having described each of the constructs, and after having established the relations to be verified in the present study, the proposed model is presented (see Fig. 1).

The five-position Likert scale, a previously validated measuring mechanism [34], was chosen for this study and a pilot test with 31 respondents was conducted to verify the tool’s performance. A number of necessary adjustments were identified and implemented, finally allowing the verification of content validity (Table 1).

The questionnaire was divided into two parts: first, a survey was conducted to profile the sample. Information on gender, age range, level of income, educational level, profession, and place and frequency of Internet access was gathered. The second part asked about each of the variables to be analyzed. This section was built with test questions measured with a 1-5 point Likert scale ranging from "Strongly disagree" to "Strongly agree"; this being the most suitable tool for measuring variables that are not directly measurable or observable [35].

Figure 1 Source: The authors

Table 1

Source: The authors

The information necessary to conduct this study was collected through an online questionnaire, given that this instrument allows for faster response, it reduces the costs of shipping and receiving the surveys, and it enables easy communication and the elimination of intermediaries [12]. Because an online shopper must be an Internet user to start with, the survey was shared via email and through social networks [36]. It was envisaged that the people invited to participate in the study were residents of metropolitan area of Medellin (Colombia) and over 14 years of age (target population).

The survey was carried out from September 16 to October 21, 2014 achieving a sample size of 300 forms, 20 of which contained discrepancies. Therefore, only 280 questionnaires were valid, making it a sample fitting the objective universe (Table 2).

The simple reliability of the measurement scales used was calculated by considering the Cronbach's Alpha values. All values above 0.70, the minimum acceptable value according to Nunnually & Bernstein [39], are shown in Table 3. The Table also shows that when applying a more accurate composite reliability analysis, the results were far superior to the 0.65 lower limit value proposed by [40].

Table 3

Source: The authors

Table 4

Source: The authors

To assess the convergent validity of the constructs, the statistical AVE (Table 3) was calculated, obtaining that all values are higher than 0.5. It was therefore verified that the items in the scales converge towards a single construct [41]. Regarding discriminant validity, the study analyzed that the correlation between the squared variables (AVE) was superior to the correlation between the constructs [42]. (see Table 4).

In order to test the predictive ability of the model a bootstrapping was performed with an N=500 subsample. From these subsamples, and based on the 0.05 significance level, all P values proved to be significant because they were under 0.05. Thus, all variables are significant and the construct is valid [43] as shown in Table 5.

Finally, with a confidence level of 95%, H1, H2, H3, H4, and H5 are significant because the P values are under 0.05 [43], as seen in Table 6.

The results show that all the hypotheses were validated as a whole (Table 6). First, a web vendor´s market orientation directly affects the social influence. This confirms the importance for a web vendor to interact with all its stakeholders online, and how this interaction will affect the actions of these groups towards such a vendor, affecting its reputation [30].

Table 5

Source: The authors

Table 6

Source: The authors

It was also verified, although to a lesser extent, that the web vendor´s customer orientation influences reputation, confirming that such strategies can enhance the vendor´s reputation [24].

Second, the relationship between a web vendor´s reputation and the risk was validated as being negative; i.e., the higher the positive reputation, the lower the perceived risk, confirming the direct relationship that other studies have shown between reputation and perceived risk [29].

Finally, it is noted that web privacy has a very important effect on reducing perceived risk. Again, an inverse relationship is presented in that as web vendors increase their privacy in terms of customer data, information management, and purchase privacy, a reduction in risk perception is obtained [7].