• No results found

The influence of product- and customer-related factors on the perceived usefulness of different types of online reviews

N/A
N/A
Protected

Academic year: 2021

Share "The influence of product- and customer-related factors on the perceived usefulness of different types of online reviews"

Copied!
51
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

June 29, 2015

The Influence of Product- and Customer-related

Factors on the Perceived Usefulness of Different

Types of Online Reviews.

Master Thesis

Author: Igor Zadorozhnyy (10825797)

University of Amsterdam Faculty of Economics and Business

(2)

2 Statement of Originality

This document is written by Student Igor Zadorozhnyy who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

Table of content

Abstract ... 4 Introduction ... 5 Literature review ... 9 Peer reviews ... 12

Advantages of peer reviews... 12

Disadvantages of peer reviews ... 13

Expert-based reviews ... 14

Advantages of expert-based reviews ... 15

Disadvantages of expert-based reviews ... 15

Factors influencing the usefulness of reviews ... 17

Review-related factors ... 18

Semantic characteristics of a review ... 18

Reviewer`s reputation ... 19

Style of a review ... 20

The number of images and words in the review ... 20

Depth and emotional extremity of a review ... 21

Product- and customer-related factors ... 21

Familiarity with a product category and frequency of a product`s usage ... 21

Type of a product ... 22

The gender of a review`s reader ... 23

Limitations of previous researches ... 23

Contributions... 25

Contribution to academics ... 25

Contribution to practitioners ... 26

Methodology ... 27

Sample ... 28

Research procedure and measures ... 29

Experiment layout ... 31

Pretest ... 31

Questions` types ... 35

Data Analysis ... 36

Results ... 38

Discussion and conclusions ... 40

References ... 44

(4)

4

Abstract

One of the methods of obtaining information about pros and cons of a product or a service is reading of reviews made either by ordinary customers (peers) or by qualified experts. Both peer and expert-based online reviews significantly influence customer`s buying behavior, which in turn influences a customer`s purchase intention (Armitage, 2003; Chen, Dhanasobhon and Smith, 2008; Harmon and Coney, 1982; Lascu, Bearden and Rose, 1995; Park et al., 2007; Petty, Cacioppo and Schumann, 1984; Zendesk, 2013). However, customers do not perceive different types of reviews as equally useful depending on different internal and external factors such as customer`s familiarity with a product category, frequency of a product`s usage, the type of a product and other product-, review- and customer-related characteristics (Cao et al. 2011; Chakravarty et al., 2010; Chen and Ho, 2015; Cheng and Ho, 2014; Ghose and Ipeirotis, 2011; Mudambi et al., 2010; Nan Hu et al., 2008; Schindler et al., 2012). The idea of this research was to find how familiarity with a product category, frequency of a product`s usage and the type of a product influence the perceived usefulness of peer- and expert-based online reviews. Having collected 102 responses from 20-35 years old Europeans, the author discovered that consumers found peer online reviews about a non-familiar not frequently used product to be more useful than peer reviews about a non-familiar frequently used product. In case of expert-based online reviews, different levels of familiarity with a product category and frequency of a product`s usage did not affect perceived usefulness of this type of reviews. The type of a product had the most significant impact on both types of reviews: consumers found peer online reviews to be more important to experience products than to feature-based ones and expert-based online reviews to be more useful to feature-based products than to experience ones. The study concludes with discussion, managerial implications and limitations which call for further research.

(5)

5

Introduction

Modern technologies (namely web-sites, online social networks and mobile apps) have made people more educated in the shopping sphere (Labrecque et al., 2013). When a person searches for a product/service online she tries to minimize the risks associated with after-purchase usage of this products/service by evaluation of its main benefits and drawbacks. A widespread technique of evaluation of the product`s pros and cons is the analysis of reviews and rankings made either by other customers or by qualified experts about this product. Although reviews are helpful in the decision making process and have a significant impact on purchase intention (Armitage, 2003; Chen, Dhanasobhon and Smith, 2008; Park et al., 2007; Petty, Cacioppo and Schumann, 1984; Zendesk, 2013), it is not clear what types of reviews (peer or expert-based) are more useful for customers and in what circumstances. Some authors state that people find expert-based reviews more reliable and, therefore, more useful than peer reviews, whereas other academics consider customer reviews to be more helpful than those with expert origin (Bearden and Etzel, 1982; Cohn and Wolfe, 2012; Harmon and Coney, 1982; Huang and Chen, 2006; Lascu, Bearden and Rose, 1995; Senecal and Nantel, 2004).

Recent literature indicates that there are several factors which influence customers` perceptions of helpfulness of a review: semantic characteristics of review, the product type, the degree of a review`s objectivity, a review`s extensiveness and emotional extremity, the reviewer`s reputation, the style of a review, the number of images and words in the review, the gender of a review`s reader, frequency of a product category`s usage, familiarity with a product category, etc. (Cao et al. 2011; Chakravarty et al., 2010; Chen and Ho, 2015; Cheng and Ho, 2014; Ghose and Ipeirotis, 2011; Mudambi et al., 2010; Nan Hu et al., 2008; Schindler et al., 2012). These studies reveal how different factors influence perceived usefulness of reviews made by customers; nevertheless, few researches were made to

(6)

6 investigate how these factors together influence customers` perceived helpfulness of not only peer reviews, but also of expert-based online reviews of a product. In other words, the influence of combinations of these factors on the perceived usefulness of expert-based reviews was not studied.

The idea of this research is to determine how such factors as the product`s type, a customers` familiarity with a product category and frequency of usage of a product category influence the usefulness of a product`s online reviews (both peer and expert-based). The purpose of this study is to find out to what degree expert-based reviews are more or less useful than peer reviews depending on the type of a product, a customers` familiarity with a product category and frequency of usage of a product category.

The influence of either the mentioned factors or different combinations of these factors on customers` perceived usefulness of peer and expert-based online reviews was poorly investigated in academic literature (Chakravarty et al., 2010; Ghose and Ipeirotis, 2011; Mudambi et al., 2010; Nan Hu et al., 2008). For example, Nan Hu et al. (2008) state that people find reviews made by a person with high reputation to be more useful. Ghose and Ipeirotis (2011) argue that based on the match between the type of a product (feature-based/experience) and reviews` degree of objectivity, the perceived usefulness of these reviews varies. Mudambi et al. (2010) state that the more extensive and the less emotionally extreme the reviewer`s comments are – the more useful the review is. Cao et al. (2011) say that the more logical and relevant a product`s review is – the more useful this review to a customer is. Also, in the study by Schindler et al. (2012) the authors found that inexpressive slang, bad grammar and spelling mistakes in a review decrease the perceived usefulness of this review. The number of images and words in a review also influence its usefulness: Cheng and Ho (2014) argue that customers perceive a wordy review with images to be better and more convincing than a brief review or a review without pictures. Bickart and Schindler

(7)

7 (2001) discovered that gender is another factor which has an effect on the perceived usefulness of a review: they state that female users of online review trust customer reviews and find them more helpful more than male users do. Finally, Chakravarty et al. (2010) discovered that depending on familiarity with a product category (movies) and frequency of usage of a product category (going to the movie theaters) peer reviews may be more useful than expert-based reviews and vice versa: those customer who are familiar with a product category and use it frequently prefer expert-based reviews to peer reviews, whereas people who are not familiar with a product category favor peer reviews. However, while mainly peer reviews were examined, expert-based online reviews did not have enough attention from the academics. Furthermore, the usefulness of both peer and expert-based reviews were discussed for services, not for products (Chakravarty et al., 2010). Finally, the research by Huang and Chen (2006) was limited to only one country and their study`s results may be not applicable to customers from other parts of the world.

The results of my research close the existing literature gap in how different factors influence perceived helpfulness of peer and expert-based online reviews. Namely, this gap will be closed after the author elicit the answer to the following research question:

“How do familiarity with a product category, frequency of usage of a product category and the type of a product together influence customers` perceived helpfulness of different types of reviews?”

As for practical implications, marketers and managers will be able to make smart decisions about introduction of the customer-based or expert-based reviewing system about its products based on the type of the product, customers` familiarity with the company`s products and frequency of usage of products.

(8)

8 In the thesis the type of a review (peer/expert-based) is an independent variable, perceived usefulness of peer and expert-based reviews on a product is the dependent variable and the type of a product (feature-based/experience)1, frequency of usage of a product category (frequent/infrequent) and familiarity (familiar/not familiar) with a product category are moderators in the “type of a review – usefulness of a review” relationship. In this research the student used experiment design and a questionnaire in order to get quantitative data

.

The sample consisted of people of 20-35 years old from Europe. Having run analysis of received responses from the sample, the author made conclusions about usefulness of both types of reviews.

In the second part of the thesis the author reveals theoretical constructs and definitions of a peer online review, an expert-based online review and their advantages and disadvantages. Also, in this part the author introduces the factors which influence perceived usefulness of a review, proposes theoretical and practical contributions and formulate the research gap, the research question and hypotheses. In the third part the author discusses the methodology he used in the research. The fourth part of the research is about analysis of the elicited data and test of the hypotheses. In the fifth part the author summarizes the results of the study and gives an answer to the research question. The sixth part of the paper is devoted to the discussion and conclusions of the research. Also, here the author explains limitations and opportunities for further research.

1 Feature-based products are the products which quality can be predicted/observed before buying the product

(these products have accurate “features”) (e.g., electronics). Experience products are the products which quality is impossible to determine upon consumption of the product (e.g., movies, books) (Ghose and Ipeirotis, 2011).

(9)

9

Literature review

In this part of the thesis the author discusses different types of reviews and how perceived usefulness of a product`s review influences buying behavior. This chapter introduces two theories of buying behavior: “Theory of Planned Behavior” (Armitage, 2003) and “Elaboration Likelihood Model” (Petty, Cacioppo and Schumann, 1984). These models help a reader to understand how cognitive and affective aspects of an advertisement about a product together with the degree of customer`s involvement into the buying process influence buying behavior of a customer. Discussion of these models leads to a conclusion that reviews and usefulness of reviews influence a person`s buying behavior. Further in this chapter the author describes two different types of reviews: peer reviews and expert-based reviews and their advantages and disadvantages. Although the author states that customers are fond of both peer and expert-based reviews, perceived usefulness of different types of reviews is not the same and depends on different factors. The next section of this literature review is devoted to the factors which affect usefulness of both types of reviews: semantic characteristics of review, the type of a product, a reviewer`s reputation, the style of a review, the number of words and images in a review, depth and emotional extremity of a review, the gender of a review`s reader, familiarity with a product category and frequency of usage of a product category. Next, the author provides a reader with several limitations of previous researches and concludes this section with the literature gap, the research question and hypotheses. In the final part of this chapter the author discusses both academic and practical contributions of the thesis.

Today customers may obtain much more knowledge about a product/service than before, primarily because of a better access to useful information (e.g.: product`s/service`s reviews, rankings, availability, prices, various types of searched product/service, etc.). Modern consumers have become more sophisticated in the process of searching, buying and

(10)

10 using of the desired product (Labrecque et al., 2013). In the “Theory of Planned Behavior” introduced by Armitage (2003) the author states that one of the variables which influences consumer behavior (purchase intention) is the consumer`s attitude toward the desired product/service. This attitude is based on both cognitive (e.g., concrete pros and cons of a product) and affective (e.g., a funny commercial about this product) responses. In the “Elaboration Likelihood Model” introduced by Petty, Cacioppo and Schumann (1984) during decision making process customers follow two possible routes: central route (a route where a customer is highly involved in the purchase of a product and generates cognitive responses) and peripheral route (a route where a customer is lowly involved in the purchase of a product and generates affective responses). In case of high involvement (interest) into a product`s purchase a consumer generates more cognitive responses than affective ones. Likewise, when a person is lowly involved (not interested) into buying something she generates rather affective than cognitive responses. This means that in order to motivate a customer to buy a product, it is necessary to find out what route the customer is on (how involved the customer is in buying the product). After the route was identified, marketing specialists have to decide what tool to use in a marketing campaign in order to make the customer generate cognitive or affective responses depending on the route. If the route and a marketing tool were identified correctly (a good fit between a route and a tool: e.g., a “central route” and “a commercial which highlights concrete pros of the product”) a customer will generate much stronger responses than if the fit between the route and the tool was weak (e.g., “central route” and “a funny commercial without concrete pros of a product”). In the “Theory of Planned Behavior” Armitage states that stronger responses refer to the higher purchase intention. Thus, when a person, for example, looks for concrete advantages and disadvantages of this product, other people`s experience of using it, some tips in where to buy this product, etc. this means that this person becomes more informed about the product and, thus, is highly involved into

(11)

11 getting this product and generates cognitive responses. Based on the discussed models, one of the tactics to make a highly involved customer be more motivated to buy a product is to strengthen the customer`s cognitive responses. This can be done, for instance, by providing a customer with accurate reviews on the product where all the advantages and disadvantages of the product, other customers` opinions about this product and personal experience in using of this product are discussed. However, if a product`s reviews are of low quality (e.g., not informative enough or too biased) or there are no any reviews at all, a highly involved customer will not be able to generate strong enough cognitive responses. This fact may negatively influence buying behavior and decrease motivation to buy a product. Thus, usefulness of reviews may influence buying behavior. The research by Zendesk, a cloud-based software help desk which enables fast growing companies with great customer service, proved that after reading positive customer reviews 42% of B2C and 62% of B2B customers purchased more (Zendesk, 2013). In case of negative reviews, 52% of B2C and 66% of B2B customers stopped buying the reviewed product or service. Totally, the researchers found out that 88% of respondents out of 1046 have been influenced by online customer reviews in the process of decision making. The study conducted by Park et al. (2007) supports the idea that useful reviews (reviews of high quality) make an impact on a customer`s decision making process.The authors indicate in their research that “reviews that are logical and persuasive, with sufficient reasons based on specific facts about the product, have a strong positive effect on purchasing intention”. In the study by Chen, Dhanasobhon and Smith (2008) the authors also found that useful reviews have a stronger impact on purchase decisions of a customer than other reviews do. The academics argue that “consumers may consider higher quality reviews more important in making purchase decisions”. Nevertheless, what type of reviews do people find more useful and why? Academics define two types of reviews: customers` (peer) reviews and expert-based reviews.

(12)

12

Peer reviews

The researchers argue that “peer reviews” is one of the most vital and practical tools for investigating the product`s pros and cons (Ling (Alice) Jiang et al., 2013). People read other customers` comments and listen to the third parties (e.g., recommender systems) when deciding what product to buy (West et al., 1999). Online customer (peer) reviews can be defined as “an important type of user-generated content, through which consumers share their experiences with products and services in order to help others make informed purchasing decisions” (Liu, Karahanna and Watson, 2011). Such reviews provide customers with useful information about quality of a product (Etzion and Awad, 2007) and reduce customers` search costs (Chen,Wu and Yoon, 2004).

Advantages of peer reviews

People like reading and making reviews for various reasons. One of the main advantages of customers` reviews is that people can experience altruistic behavior during the reviews` making process (Hennig-Thurau, Gwinner, Walsh and Gremler, 2004). In this case, “altruism refers to consumers sharing their positive or negative experiences to assist other consumers in making buying decisions”. People make reviews because they want to help other customers with buying decisions or to warn others about negative experience, or both (Engel et al., 1993). Moreover, satisfied customers make positive reviews about a company or a product in order to help the company; in other words, they want to give something in return for great experience (Sundaram et al., 1998). Thus, user-generated reviews are a perfect tool to express customers` strong feelings about a product or a company.

Another benefit of peer reviews is the economic reward for making reviews. Hennig-Thurau et al. (2004) state that companies encourage customers to make and share positive content (e.g., reviews, video ads, pictures, etc.) by offering a cash prize. For instance in 2006, Frito-Lay offered one of five US$10,000 cash prizes to those who will create the best

(13)

30-13 second Doritos ad to be shown during the 2007 Super Bowl (SuperBowlAds, 2007). As a result, the winning ad became the best-liked spot in the Super Bowl 2007. Thus, the winner got the money prize and Frito-Lay made an enormous propaganda and improved its image.

In the work by Park et al. (2007) the authors state that, contrary to expert-based reviews, customer reviews provide people with experience oriented information about a product. Consumers find such reviews as representation of the other people`s real experience from a product usage. This fact makes customer reviews more believable and understandable than expert-based reviews.

Finally, online reviews offer social benefits for customers. Exchanging thoughts about a product or a company facilitates online communication between customers (Bickart and Schindler, 2001). The authors argue that participation in online forums, newsgroups, listservs and bulletin boards can significantly influence consumer behavior. In consumer communities people can share their opinions, obtain product information and develop relationships with those who share same interests. What is more, membership in online consumer community may not only facilitate participation in community-related activities, but also lead to increased product purchases.

Disadvantages of peer reviews

Researchers Liu, Karahanna and Watson (2011) in their work state that a big number of reviews may increase a customer`s search costs. This happens when large quantity of evaluations prevent a customer from sifting through the reviews. Moreover, conflicting opinions can puzzle a person to make a weighted choice. The overload of number of reviews may lead to pre-choice confusion, stress and anticipated regret (Krähmer, 2013; Lipowski, 1970), as well as, to post-choice uncertainty about choice and regret (Bell, 1982).

Reviews on a product or service may be very subjective: for example, in a process of evaluating a restaurant on a special website a person states that she liked service and food

(14)

14 quality, but disliked too noisy background music (puts “4” out of “5” in a rating list). Another person says that she liked service too, but was disappointed that she could not pay credit card there (puts “3” out of “5”). After these reviews have been made the average grade is 3.5/5. Nevertheless, there are customers (call them “the third group”) who are fond of loud music in a café/restaurant and prefer paying cash for their meal. Consequently, even though the food quality and service are great, the entire grade for this restaurant is not high. This fact will probably persuade the third group of customers not to go there. In this context, peer reviewing may not be the perfect instrument of evaluation of a product or a service.

Furthermore, Salganik (2008) in his paper argues that peer reviewing may be useless because of such social phenomenon as herding. Huang and Chen (2006) define herding as a behavior when “consumers monitor the comments of others regarding specific topics and use them as a basis for their own choices”. In the study held by Salganik (2008) the list of songs on a special web-site where people can download songs was inverted in a way that the popularity rating of songs was turned upside-down. So, the first popular song became the 48th and the last popular song became the first one. Consequently, the number of downloads changed in the inverted way because people tend to think that the best/coolest song is the one which has the largest number of downloads. The participants` choices of downloading were based not on their musical preferences but on the other people`s choices of downloading. In this case, Salganik illustrated how irrational people may be when using only peer review evaluation system. When peer reviews make little use, expert-based reviews may become an alternative to reviews of other customers.

Expert-based reviews

Similar to peer reviews, expert-based reviews are made in order to evaluate a product`s features but with help of real experts – people who can provide the customer with

(15)

15 the accurate quality evaluation of the searched product. Expert-based review is a review that comes along with the rules of honesty and objectivity2.

Advantages of expert-based reviews

One of the advantages of expert-based reviewing over peer reviewing is that generally expert-based reviews are more objective than peer reviews and do not depend on other people`s attitudes toward the product/service. Crisci and Kassinove (1973) state that advice`s strength and perceived level of expertise have a positive influence on a person`s compliance with recommendations of the source.

What is more, expert-based reviews will make the process of the product`s search easier by reducing consumers` searching costs (e.g.: time, emotional costs associated with lack/excess of a product`s reviews). Lower costs of online search of a product may decline consumers' price sensitivity by highlighting brands` differentiation (Yadav and Pavlou, 2014).

Also, in the work by Katherine N. Lemon et al. (2002) the authors argue that consumer`s future considerations about a particular product/service (namely, anticipated benefits and anticipated regret) influence the customer`s satisfaction. Expert-based reviewing will maximize anticipated benefits and minimize anticipated regret associated with a product by providing customers with qualified opinions about a product/service, thus, enhancing customers` satisfaction.

Disadvantages of expert-based reviews

Kambil and van Heck (2002) state that expert-based reviews are less helpful than peer reviews. They found that small groups of elites frequently perform worse than large groups of people in the processes of problem solving and even in making predictions about future. The

(16)

16 authors explain this fact that regardless of expert`s knowledge, experts possess only limited amounts of data.

Secondly, Huang and Chen (2006) in their study criticize expert-based reviews for the fact that experts may also be biased. Like an every person, an expert has her own opinion about a particular issue. Bias in an expert`s opinion may occur, for example, when she writes a negative review on a restaurant not because of bad food quality or service, but because she has a poor relationship with the owner of the restaurant. Also, an expert may be bribed by a person or a company to write a negative review about its competitor`s products. In this case, the expert`s objectivity is under question.

All in all, Huang and Chen (2006) say that some people prefer peer online reviews while other customers find expert-based online reviews more useful. In different cases people find reviews more or less helpful, depending on the context (Chakravarty et al., 2010; Ghose and Ipeirotis, 2011; Mudambi et al., 2010; Nan Hu et al., 2008). The authors explain this phenomenon by saying: “various online recommendations may influence consumer choices in different ways because consumers may consider them to have varying degrees of credibility”. For some customers expert-based reviews are more important: recommendations from people with high-credibility are likely to be accepted (Bearden and Etzel, 1982). Furthermore, trustworthiness and source expertise have a positive influence on the consumers` attitudes toward a brand, their intentions, and purchase behaviors (Harmon and Coney, 1982; Lascu, Bearden and Rose, 1995). With the use of expert-based reviewing system a company will provide customers with the comprehensive and transparent evaluation of its products, a fact which will improve customers` loyalty and trust toward this company (Cohn and Wolfe, 2012). Thus, different studies show that expert-based reviews influence consumer product choices relatively more effectively than peer reviews do. On the other hand, several studies indicate that people favor less expert-based reviews and more reviews of other customers.

(17)

17 Senecal and Nantel (2004) argue that in the decision making process, in comparison with experts, “other consumers are considered a more trustworthy source of recommendations”. In the study by Huang and Chen (2006) the authors found that peer online reviews have a bigger impact on consumer choices than those with expert base. In the experiment conducted by the authors, respondents found recommendations of their peers more reliable than reviews from professional critics during decision making process. Although the authors propose that peer reviews are more helpful than expert-based ones only because “crowds are right more often than experts are”, it is unclear from this study what other factors influence customer`s attitude to peer and expert-based online reviews in their decision making process. In the following section the student covered various factors which have an impact on the perceived usefulness of different types of reviews.

Factors influencing the usefulness of reviews

This section discusses different factors which influence perceived helpfulness of a review. Among these factors are semantic characteristics of review, the type of a product, a reviewer`s reputation, the style of a review, the number of words and images in a review, depth and emotional extremity of a review, the gender of a review`s reader, familiarity with a product category and frequency of usage of a product category. In order to make a clearer picture the author allocated the mentioned factors into two groups:

• Review-related factors:

o Semantic characteristics of review; o Reviewer`s reputation;

o Style of a review;

o Number of words and images in a review; o Depth and emotional extremity of a review.

(18)

18 • Product- and customer-related factors:

o Familiarity with a product category; o Frequency of usage of a product category; o Type of a product;

o The gender of a review`s reader.

The author chose to investigate how product- and customer-related factors influence the perceived usefulness of reviews, because, in comparison with well-examined factors from review-related group, product- and customer-related ones were examined less prominently. What is more, having analyzed how product- and customer-related factors influence the usefulness of reviews one may conclude what reviewing system a marketer should implement in her business. For example, the marketer`s decision about what reviewing system to use may change depending on customers` familiarity with a product category (familiar/non-familiar), frequency of usage of a product (frequently/not frequently), type of a product (feature-based/experience) or the gender of a review`s reader (male/female). In comparison with product- and customer-related factors, review-related factors may be difficult to determine in advance. For instance, a marketer can hardly predict what style a customer will use in her review, how many words and images there will be in a review, how deep the review will be, etc. Thus, discovery of the impact of product- and customer-related factors on perceived usefulness of reviews makes more sense for practical contributions. The researches` findings have several limitations which make a base for a further research.

Review-related factors

Semantic characteristics of a review

The research by Cao et al. (2011) indicates that one of the most important factors which influence the perceived usefulness of a review is semantic characteristics of a review. The authors argue that semantic characteristics of a review (the substance of a review) play a

(19)

19 major role in the customer`s evaluation of a review`s usefulness: for example, customers will find a review “Product “A” is good, I like it.” not useful since this review lacks the line of reasoning of what exactly this person liked product “A” for. It should be mentioned here that the results of the study show semantic characteristics of a review to have a stronger effect on the perceived usefulness of a review than such factors as basic characteristics of a review. Such characteristics include pieces of information about whether a review discusses pros and cons of the product, has a reviewer`s summary of her experience in using of the product, how many days since the posting date, the extremeness of the review (see factor “depth and emotional extremity of a review”) and stylistic characteristics of a review (the average number of words in the sentence, the average number of characters per word, etc.) (Cao et al., 2011).

Reviewer`s reputation

Nan Hu et al. (2008) say in their work that “trust and reputation leads to reduced behavioral uncertainty and decreased transaction costs because trust in a contractual relationship can result in more accurate and timely exchange of information and greater influence on the information receiver”. When a person looks up for a product`s reviews, the perceived trust to the source of the review plays an important role in the decision making process. The authors state that consumers prefer reviews made by reviewers with high reputation because people find such reviews as trustworthy and credible. Moreover, Cheng and Ho (2014) in their study discovered that reviews made by a person with a higher level of expertise are perceived as more useful. Trust from customers may be enhanced, for example, if the reviews of a product/service are made by people with high reputation and whose opinions are reliable and objective. For example, the company may introduce expert-based reviewing system about its products in order to increase its reputation and customers` trust.

(20)

20 Style of a review

In the paper by Schindler et al. (2012) the researchers tested whether the style of a review influence the helpfulness of a review. They defined two groups of the factors:

• Style factors related to weaker impact: misspelling, bad grammar, the use of inexpressive slang, the use of qualifications and repetition and;

• Style factors related to stronger impact: expressive slang, humor, good grammar, presence of personal information and first-person pronouns, self-effacing wording.

The authors proposed that presence of negative style characteristics in a review would be associated with reviews of low value, whereas presence of positive style characteristics – with high value reviews. The results of the research are controversial: negative style characteristics decreased the value of a review; however, positive style characteristics did not necessarily increase a review`s value. In case of the latter characteristics, the authors stated that some use of positive characteristics insignificantly increase the value of a review, but overuse of such characteristics detracts a review`s value. Thus, in order to have the highest review`s value, a reviewer should not use negative style characteristics and avoid overuse of positive ones.

The number of images and words in the review

In the study by Cheng and Ho (2014) the authors argue that a review`s usefulness depends on the number of words and images in it too. However, the effect of the number of images on the perceived usefulness of a review is stronger than of the number of words in this review. The academics conducted a study where they examined the perceived usefulness of reviews of Chinese restaurants. They found out that a wordy review is perceived to be more useful than a brief review. Cheng and Ho discovered that if a review about a restaurant

(21)

21 lacks any images (e.g., food images) but the reviewer adds images to her review, the usefulness of this review increases dramatically.

Depth and emotional extremity of a review

Mudambi et al. (2010) state that the review`s depth (the extensiveness of the reviewer comments) about the product has a positive effect on the helpfulness of the review. Moreover, the authors found out that for experience goods, reviews with moderate ratings are more helpful than with extreme ratings. Based on the assumption that, in comparison with peer reviews, expert-based reviews are generally more extensive and less emotionally extreme (more moderate), this research`s purpose is to find out whether expert-based reviews are more helpful than peer reviews depending on different circumstances.

Product- and customer-related factors

Familiarity with a product category and frequency of a product`s usage

A concept of frequency of a product purchase and familiarity with a product category also has an impact on the customers` attitude toward helpfulness of reviews. In the paper by Chakravarty et al. (2010) the researchers found that word-of-mouth (peer reviews) and critical (expert-based) reviews differently influence people`s evaluation of the movie depending on their frequency of buying a product, namely going to the cinema. The authors state that critical reviews have a greater influence on frequent moviegoers` attitude toward a film than mouth, while infrequent moviegoers are influenced more by word-of-mouth than by critical reviews. The authors` conclusions are academically interesting in an extent that frequency of usage of a product and familiarity with a product category have an impact on perceived helpfulness of different types of reviews. Although the academics explored only one product (movies), other products should be examined more closely. Based on the findings of Chakravarty et al. (2010), the author developed the following hypotheses:

(22)

22 H1a.Expert-based online reviews are more useful for customers who are familiar with the product category.

H1b.Peer online reviews are more useful for customers who are not familiar with the product category.

H2a.Expert-based online reviews are more useful for customers who use the product category frequently.

H2b.Peer online reviews are more useful for customers who do not use the product category frequently.

Type of a product

In the study by Ghose and Ipeirotis (2011) the authors discovered that reviews which contain both objective and highly subjective sentences negatively influence the product sales, in comparison with the reviews that include only objective or only subjective information. Moreover, their research revealed that customers find a review which contains a mixture of both objective and subjective elements more helpful than a totally objective or subjective review. It should be mentioned here that depending on the type of a product people prefer more objective or more subjective reviews. For example, the author discussed that in case of feature-based products (like electronics) customers like reviews which consist of far more objective than subjective elements (generally confirm the product description`s validity). On the other hand, for experience products (e.g., a movie) people are more interested in the subjective reviews (personalized sentimental aspects of the film) than in the objective reviews (e.g., the film`s plot). As was mentioned above, expert-based reviews are generally more objective than peer reviews. To this extent it is interesting to investigate whether peer or expert-based reviews about a feature-based/experience product are more useful. Thus, the author came up with the following hypothesis:

(23)

23 H3a.Expert-based online reviews are more helpful in the process of purchase of a feature-based product than an experience product.

H3b. Peer online reviews are more helpful in the process of purchase of an experience product than a feature-based product.

The gender of a review`s reader

Bickart and Schindler (2001) analyzed what sources of information about a product (internet-forum information or marketer-generated online information) different customers (male/female) perceive to be more useful. The authors define internet-forums as electronic platforms where customers can exchange their impressions about products, give pieces of advice to other customers about products and discuss pros and cons of products. Therefore, such internet-forums play a role of a site where customers exchange user-generated content or, in other words, their reviews. Bickart and Schindler (2001) found that female users found information about products elicited from internet-forums to be more reliable than male users did. Thus, one may make an assumption that, in comparison with male users, female users perceive peer online reviews to be more helpful in the process of getting information about products. Furthermore, the authors discovered that internet-forums generate much more product interest than corporate-websites do.

Limitations of previous researches

Although academics identified several factors which influence perceived usefulness of different types of reviews, there are several limitations in their studies. Firstly, in the study by Ghose and Ipeirotis (2011) about how objectivity and subjectivity influence perceived usefulness of a review, one of the limitations to the authors` work is that their research was focused only on one e-commerce retailer (Ghose and Ipeirotis, 2011). In order to find out whether objective/subjective reviews generally are more appropriate for

(24)

feature-24 based/experience products a researcher should examine different industries and retailers. Secondly, although the Chakravarty`s et al. (2010) study is about people`s attitude toward movies, a further research may be focused on familiarity with a product category and frequency of usage of other product categories (Chakravarty`s et al., 2010). Thirdly, Mudambi et al. (2010) discovered how emotional extremity and depth of a review moderates the usefulness of a review; however, they analyzed only six products (MP3 player, laser printer, music CD, PC video game, digital camera and cell phone) in their study, so broader set of products/services should be examined more closely (Mudambi et al., 2010). Fourthly, Huang and Chen (2006) conducted their study in Taiwan and examined the local respondents` attitude toward helpfulness of different types of reviews about only one product (the book) (Huang and Chen, 2006). In order to make the authors` findings reliable, one should conduct another study which will cover other countries and other product/service categories. Finally, impact of different combinations of the mentioned factors on the perceived usefulness of different types of reviews was not studied before.

According to the studied literature, there is a literature gap in determining usefulness of products` peer and expert-based online reviews: it is little known how familiarity with a product category, frequency of usage of a product category and the type of a product taken together influence customers` attitude toward helpfulness of different types of reviews.

Having defined the literature gap, the author came up with the following research question:

“How do familiarity with a product category, frequency of usage of a product category and the type of a product together influence customers` perceived helpfulness of different types of reviews?”

In order to answer the research question, the author proposed the following hypotheses.

(25)

25 Hypotheses:

H1a. Expert-based online reviews are more useful for customers who are familiar with the product category.

H1b. Peer online reviews are more useful for customers who are not familiar with the product category.

H2a. Expert-based online reviews are more useful for customers who use the product category frequently.

H2b. Peer online reviews are more useful for customers who do not use the product category frequently.

H3a. Expert-based online reviews are more helpful in the process of purchase of a feature-based product than an experience product.

H3b. Peer online reviews are more helpful in the process of purchase of an experience product than a feature-based product.

Contributions

In this part of the thesis contributions both to academics and to practitioners are discussed.

Contribution to academics

This research provides academics with useful findings in determination of perceived usefulness of different types of reviews which is based not only on the discussed factors, but also on different combinations of these factors. Moreover, helpfulness of different types of reviews of products was poorly investigated. In this paper the student plans to explore the usefulness of peer and expert-based online reviews depending on such factors as customer`s familiarity with a product category, frequency of usage of a product category and the type of a product and to close the existing literature gap.

(26)

26 Contribution to practitioners

This paper makes practical implications for managers too. The results of the research will make it clear whether the mentioned factors influence helpfulness of peer and expert-based online reviews to customers. Consequently, marketers and managers will be able to make smart decisions about introduction of the peer-based or expert-based reviewing system about their products based on customers` familiarity with the company`s products, frequency of usage of its products and the type of the product.

In this chapter of the thesis the author explained what role reviews play in the concept of buying behavior, discussed two different types of reviews and explained their pros and cons. Moreover, the student revealed that depending on different factors peer reviews and expert-based reviews differ in usefulness. In the next section the factors were discussed in details. Then, the author discovered limitations of previous studies and came up with the literature gap, the research question and hypotheses. Finally, academic and practical contributions were discussed.

(27)

27

Methodology

In the previous chapter the author discovered that perceived usefulness of a review makes a significant impact on a customer`s buying behavior. In order to determine how different factors (such as type of a product, frequency of a product usage and familiarity with a product category) influence the perceived usefulness of different types of reviews, the author conducted a study. In the following chapter the research design is discussed. The author clarifies what dependent variables, independent variables and moderators in this variables` interaction are and introduces a conceptual model of his study. Also, the author explains what research techniques he used during his research, describes the sample, the experiment layout and the types of questions he used in the research.

In this research the type of a review (peer/expert-based) is an independent variable, perceived usefulness of an expert-based review on this product is a dependent variable and the type of a product (feature-based/experience), frequency of usage of a product (frequent/infrequent) and familiarity (familiar/not familiar) with a product category are moderators on the relationship “type of a review – usefulness of a review”. In the previous chapter the author found that perceived usefulness of a review has a significant impact on buying behavior, which in turn influences a customer`s purchase intention (Armitage, 2003; Chen, Dhanasobhon and Smith, 2008; Harmon and Coney, 1982; Lascu, Bearden and Rose, 1995; Park et al., 2007; Petty, Cacioppo and Schumann, 1984; Zendesk, 2013). The interaction between the independent variable, the dependent variable and moderators is depicted in the conceptual model below.

(28)

28 Picture 1. The conceptual model of influence of the factors on perceived usefulness of a review in the “type of a review – usefulness of a review” relationship

Sample

The author used probability sampling in his research because of the following reasons:

• the chance of each respondent being selected from the population is known and is equal for all cases;

• the research question requires to estimate statistically the peculiarities of the population from the sample;

• the author plans to use experimental research strategy.

The author used stratified random sampling because he needed to get data only from specified strata: young people from Europe. Because the research question is about perceived usefulness of online reviews, the sample consisted of 20-35 years old people as customers from this age group are the most frequent users of internet3 and online shopping4. Because in this research the gender of respondents is less important than the age and nationality of respondents the author treated this respondents` characteristic as a control variable. The online questionnaire was sent by e-mails to participants and posted on social networks. The survey was conducted in the end of March and lasted till the end of April.

3

http://www.statista.com/statistics/272365/age-distribution-of-internet-users-worldwide/

(29)

29 Research procedure and measures

As the goal of this paper was to find how independent variable and moderators influence dependent variable the student used an experimental research strategy. Because the author needed to elicit answers to close questions from big number of respondents and in a short period of time (from two to six weeks) he used a questionnaire as a technique of data collection. The experiment consisted of six different mini-cases about different products and one peer- and one expert-based online review on a product in each case. These mini-cases differed from each other by the object of analysis: each mini-case discussed different products which varied in terms of their type, consumers` familiarity with them and frequency of their usage. The author assumed that depending on the category each product fit, the perceived usefulness of the reviews on this product varied:

Mini-case 1 (Factors 1, 2 and 3: the type of a product + familiarity with a product category + frequency of usage of a product category): “familiar frequently used feature-based product – usefulness of reviews”;

Mini-case 2: “familiar not frequently used feature-based product – usefulness of reviews”; Mini-case 3: “familiar frequently used experience product – usefulness of reviews”; Mini-case 4: “familiar not frequently used experience product – usefulness of reviews”;

Mini-case 5: “non-familiar not frequently used feature-based product – usefulness of reviews”;

Mini-case 6: “non-familiar not frequently used experience product – usefulness of reviews”; Note: two of the possible eight interactions between “the type of a product” (feature-based/experience), “familiarity with a product category” (familiar/non-familiar) and “frequency of usage of a product category” (frequently/not frequently) were not discussed. Namely, it is senseless to study the interactions “non-familiar with a product – frequently

(30)

30 used product” because if a person experiences low degree of familiarity with a product category it automatically means that she does not use this product frequently (or never used a product). Thus, the interactions “non-familiar frequently used feature-based product” and “non-familiar frequently used experience product” were not discussed.

The technique of determining the usefulness of reviews was adapted from the study of the influence of reviews` usefulness on customers` decisions by Connors, L., Mudambi, S. and Schuff, D. (2011). In order to discover how useful respondents find different types of reviews in each case participants had to assign a value to a degree of peer and expert-based online reviews` usefulness from “1 – not useful” to “4 – very useful”. Thus, the categorical data (perceived level of usefulness) were converted into numerical data: “not useful” = 1, “barely useful” = 2, “useful” = 3 and “very useful” = 4. The independent variable “the type of a review” is categorical. Based on the assumption that the independent variable is categorical and the dependent variable is numerical the author decided to use analysis of variance (ANOVA) as a method of statistical data analysis. The student used SPSS to conduct data analysis. The logic of the author`s choice is depicted in the picture below.

Picture 2. Methods of data analysis depending on the type of variables

(31)

31 Experiment layout

The cases differed from each other by the object of analysis or, in other words, by the product. When a participant clicked the link, the window with the survey description appeared. After a respondent read the survey description, the survey started. The first two questions required a respondent to indicate his/her age and nationality. Having inserted his/her age and nationality a participant moved to the first mini-case.

Each product was introduced with two reviews on this product: one peer online review and one expert-based online review. Having finished with the reviews` reading, participants needed to evaluate how useful they found each review by assigning a value from “1 – not useful” till “4 – very useful”. When a respondent ended with the reviews` evaluation she moved to the next mini-case. Having answered the questions to the sixth mini-case, respondents finished the survey. One may find the example of a question from the survey in Appendix E. Before running the survey, the author needed to find out what products to include in the study. In order to find the right product for analysis the author made a pretest.

Pretest

In order to determine what product fits better to each interaction, the author made a pretest, which consisted of a survey. The survey required participants to answer the questions about familiarity with a product and frequency of usage of a product. Totally, there were three products in each category to evaluate. Based on the respondents` answers the author had to choose one product which fitted each interaction better than other two alternatives. In Appendix A one may find the table with all the alternatives. Below the explanation of the products` choice for pretesting procedure is discussed.

(32)

32 As was mentioned before, feature-based products are those products which characteristics consumers can know in advance, before consumption. For example, mobile phone, laptop or printer – a person clearly knows what to expect from the product he or she bought (e.g., for a mobile phone or a laptop it is the processor capacity or display resolution, for a printer – is the printing capacity and paint consumption). The same principle can be applied to other feature-based products: a capsuled coffee-machine (the volume of coffee output or types of coffee capsules supported), a car (fuel consumption rate, engine power) and an air-conditioning (the power of the fan, energy consumption rate). As for experience products, an MP3-player is a product which main feature, sound quality, a consumer cannot know upon consumption. Although, there is a standard characteristic of an MP3-player which a person can know before consumption – memory volume (e.g., 8 gb or 64 gb), it is not the main feature of an MP3-player that customers are interested in. Therefore, one may assume that an MP3-player is rather experience product than a feature-based one. The same logic was applied for a photo-camera: consumers can never know whether the camera is good before they see the actual pictures it takes. Probably the borders between absolutely feature-based or absolutely experience products are not clear enough. However, if one looks at these products he or she may say that a particular product is more experience than feature-based and vice versa.

In order to determine the frequency of usage of a product the student used an adapted framework from the study by Hamilton et al. (2011). The authors asked the following question to the participants: “How frequently [do you play video games/would you use a scientific calculator if you had one]?” The participants had to answer this question by choosing one of the possible alternatives: “Less than once a year”, At least once a year”, “At least once a month”, “At least once a week”, “At least once every three days” and “At least once a day”. The student assigned values to each answer: “Less than once a year” = 1, “At

(33)

33 least once a day” = 6. Thus, a product with the highest mean value was supposed to be the most frequently used, whereas a product with the lowest mean value – the least frequently used. Based on the nature of the interaction the student chose the best fitted product.

As for familiarity with a product, the author chose the questionnaire from the study by Dube and Schmitt (1999). Academics asked the survey`s participants to assign a value to the degree of familiarity with a particular product: they asked to assign the number from 1 – “not familiar at all” to 4 – “very familiar” to every discussed product. Thus, as with frequency of a product`s usage, a product with the highest mean value was perceived as the most familiar from consumers` point of view, whereas a product with the lowest mean value – the least familiar. The samples with questions about frequency of a product`s usage and familiarity with a product can be found in the Appendices B and C respectively. Totally, the author received 27 answers. The logic for choice of products which fitted each interaction best is depicted below.

Having obtained the answers from respondents, the student counted the mean value for the levels of familiarity with a product category and frequency of its usage of all participants. Next, depending on the conditions of each interaction, the author chose the most appropriate values. For example, in the interaction “familiar frequently used feature-based product” (the first group of products: mobile phone, laptop, printer) the student picked the maximum value of the frequency of a product`s usage and the maximum value of familiarity with a product category. In other words, the author looked for the products with the highest levels of familiarity and frequency. Having calculated the sums of these values for every product, the student picked the one with the biggest sum: a mobile phone (9.76 out of 10) in the “familiar frequently used feature-based product” interaction and an MP3-player (7.75 out of 10) in the “familiar frequently used experience product”. The same technique was used for non-familiar non-frequently used products: the author picked the lowest value of “frequency

(34)

34 + familiarity” sum: a fishing rod (1.40 out of 10) in the “non-familiar not frequently used feature-based product interaction and a smartwatch (2.40 out of 10) in the “non-familiar not frequently used experience product” interaction. As for the products in mixed categories (familiar but not frequently products), the author calculated standard deviation for every value. If standard deviation was small (less than 0.5), the student assumed that there was no difference in what product to choose. In case standard deviation was big (more than 0.5), the author picked the product with the lowest or highest value, depending on the specified criterion (interaction). Having determined the right products, the student turned to the main survey. The defined products can be found in the Table 1.

Table 1. The products with the best fit to each interaction

Mini-case Interaction Product

1 Familiar frequently used feature-based product Mobile phone 2 Familiar not frequently used feature-based product Air conditioner 3 Familiar frequently used experience product MP3 player

4 Familiar not frequently used experience product Circus/theater performance 5 Non-familiar not frequently used feature-based product Fishing rod

6 Non-familiar not frequently used experience product Smartwatch

In the survey all cases started with the introduction of a product which fit one of the possible six combinations of factors discovered earlier. Each mini-case required a respondent to answer the following question: “How useful do you find each review on this product? Assign a value from 1 (not useful at all) to 4 (very useful)”.

These mini-cases 1 and 2 the following structures:

1. A product is introduced. The mini-cases 1 and 2 started with introduction of two familiar feature-based products which vary in degree of frequency of their usage.

2. Reviews on this products are introduced. Then, participants read four reviews about each product (one peer review and one expert-based review on a

(35)

35 familiar frequently used feature-based product; and one peer review and one expert-based review on a familiar not frequently used feature-based product). 3. Usefulness of the review is evaluated. Finally, participants assigned a value

to the usefulness of these reviews based on a four-scale metrics (not useful at all – extremely useful).

After one type of a product was studied (a feature-based), another type of a product (experience) was introduced and the mini-cases 3 and 4 started. Two familiar experience products were introduced which varied in the degree of participants` frequency of these products` usage. Participants repeated the steps 2 and 3 but with another type of a product.

Finally, respondents moved to the mini-cases 5 and 6. Here, two types of non-familiar not frequently used products (feature-based and experience) were discussed. Participants repeated the steps 2 and 3 and ended the survey.

Questions` types

In this survey different types of questions were used: rating questions, an open question and a quantity question. The author introduced rating questions in order to get respondents` evaluation of each review. Respondents had to assign a value to the usefulness of both peer- and expert-based reviews with the use of a four-scale ranking: “not useful”, “barely useful”, “useful” and “very useful” (this technique was adapted from the study by Connors, L., Mudambi, S. and Schuff, D., 2011). An open question required a person to indicate his/her nationality, when a quantity question – the respondent`s age. Both open and quantity questions appear only once in the survey.

(36)

36

Data Analysis

In this part the student discusses the process of data analysis. Namely, he reveals the results of missing data analysis, computing of counter-indicative items and the logic of choice of a regression model for further data analysis. Then, the author describes the final process of data analysis and introduces results.

Totally, the author elicited data from 102 respondents. The mean age was 24 years, standard deviation was 2.5 years. The youngest and the oldest participants were of 20 and 33 years old respectively. Because there were six different mini-cases about six distinctive products each participant had to evaluate the usefulness of both types of reviews for all six mini-cases. Therefore, totally the author obtained 612 answers from 102 people. Each respondent answered all the questions in the questionnaire. No missing values were discovered during the analysis of missing data. Because respondents had to assign values to the degree of how useful they found each review on each product, the author used 4-point Likert scale: from “1 – not useful” to “4 – very useful”. Therefore, those who put the higher value found the review more useful. Contrary, reviews with lower values were perceived as less useful. Thus, there were no counter-indicative items in this research. The survey consisted of only one question: "How useful do you find a peer review and an expert review on the product N?" N: [1...6]. Therefore, the dependent variable was composed of only one item (“Review`s Usefulness” variable). There was no need to create any new scales in order to find the usefulness of the reviews. Taken into account that the dependent variable was composed of only one item, the author did not need to run EFA/Reliability analysis.

The author conducted three separate analyses in order to find how the mentioned factors affected perceived usefulness of peer and expert online reviews. Each moderator had

(37)

37 two levels: familiarity with a product category: familiar/not familiar; frequency of a product`s usage: frequently/not frequently, and the type of a product: feature-based/experience.

Because the independent variable (the factor) is categorical and the dependent variable (the usefulness of reviews) is numerical, the student ran two-sample t-test for all three analyses. The first analysis was about how familiarity with a product category influenced usefulness of peer and expert online reviews. In the second analysis the effect of frequency of a product`s usage over the usefulness of reviews was studied. The third analysis was run in order to find out how different types of product impact the reviews` usefulness. The results of the survey are depicted in the Table 2.

Table 2. Group Statistics "Familiarity, Frequency, Type of a product – Usefulness of Reviews"

Familiarity Mean SD

Peer Usefulness High 2.87 .795 Low 3.18 .694 Expert Usefulness High 3.12 .741 Low 3.13 .661

Frequency Mean SD

Peer Usefulness High 2.83 .803 Low 3.05 .752 Expert Usefulness High 3.11 .744 Low 3.12 .701

Type of a product Mean SD

Peer Usefulness Feature-based 2.82 .712 Experience 3.12 .808 Expert Usefulness Feature-based 3.20 .717 Experience 3.04 .705

(38)

38

Results

In this section, the author describes the results of the whole study. Based on the chosen statistical model, he makes conclusions about the settled hypotheses. Furthermore, the student reveals what was the most and the least influential factor on the perceived usefulness of both types of reviews what type of reviews was perceived as more useful.

The first analysis shows that different levels of familiarity with a product category have a significant negative effect on the perceived usefulness of peer reviews: t (610) = – 4.646, p < 0.001 (two-tailed) and non-significant effect on the usefulness of expert reviews: t (610) = – 0.200, p = 0.842 (two-tailed). One may find the results of all three two-sample t-tests in the Appendix D. Therefore, people who were not familiar with a product category found peer online reviews about this product more useful than peer online reviews about a product category people were familiar with. Thus, H1.b was confirmed. In case of both high and low familiarity with a product category, there was almost no difference between the perceived usefulness of expert online reviews: M = 3.12, SD = 0.741 for high familiarity and M = 3.13, SD = 0.661 for low familiarity. Thus, H1.a was rejected.

In the second analysis, frequency of a product`s usage influenced the perceived

usefulness of peer online reviews also significantly and negatively: t (610) = – 3.306, p < 0.01 (two-tailed), but usefulness of expert online reviews – not significantly: t (610) = – 0.160, p = 0.873 (two-tailed). People who did not use a product frequently,

favored peer online reviews about such products more than peer reviews about a product which respondents used frequently. Thus, H2.b was confirmed. Different levels of frequency of a product`s usage did not change peoples` attitude toward expert online reviews: M = 3.11, SD = 0.744 for high frequency and M = 3.12, SD = 0.701 for low frequency. Thus, H2.a was rejected.

(39)

39 In the third analysis, different types of a product had a significant effect on both types of reviews: a negative effect on peer online reviews: t (610) = – 4.883, p < 0.001 (two-tailed) and a positive effect on expert online reviews: t (610) = 2.786, p < 0.05 (two-tailed). The research showed that people perceived peer online reviews about experience products as more useful than those about feature-based ones. On the other hand, respondents demonstrated that expert online reviews about feature-based products were more useful than those about experience products. Thus, H3.a and H3.b were confirmed.

Finally, various values of “t” in the two-sample t-tests indicate that the degree of factors` influence differ. The author found that the type of a product is the most influential factor over perceived usefulness of reviews: t (610) = – 4.883, p < 0.001 (peer) and t (610) = 2.786, p < 0.05 (expert). Familiarity with a product`s category is less influential than the type of a product: t (610) = – 4.646, p < 0.001 (peer), for expert reviews – not significant: p = 0.842. The least influential factor is frequency of a product`s usage: t (610) = – 3.306, p < 0.01 (peer), for expert reviews – not significant: p = 0.873.

Apart from results elicited from the hypotheses testing, the student made another interesting conclusion about usefulness of peer- and expert-based online reviews. Having calculated the means for usefulness of different types of reviews in each analyses, the author found that expert-based online reviews were always perceived more useful than peer online reviews: in the first analysis about familiarity with a product category the usefulness of peer online reviews and expert-based online reviews is M = 3.03 and M = 3.13 respectively. In the analysis of frequency of a product`s usage the similar values for peer and expert reviews are M = 2.94 and M = 3.12 respectively. Finally, in the third analysis of the type of a product the usefulness of peer reviews is M = 2.97 and of expert reviews is M = 3.12. Therefore, expert-based online reviews are generally more useful than peer online reviews.

Referenties

GERELATEERDE DOCUMENTEN

In addition, we show how the translation can be extended beyond the basic flow graph model and safety logic to richer behavioural models (such as open programs) and richer

Methods Design Design Design Design Teaching program Student Student’ Teacher Education Methods Compe- tences Classic Classic Entrepreneurial Entrepreneurial Culture..

grow rapidly than the demand for primary commodities, particularly agricultural products. This means that the growth in demand for the exports of LDC is unlikely

Conceptual model Central content Lexical Complexity Two-sidedness Peripheral cues Aggregated rating score volume Review helpfulness volume Reviewer’s expertise Real

From this research it can be concluded that there are no significant differences between humorous and non-humorous reviews, and no significant differences between

Since positive reviews for search goods help to increase a consumer’s product attitude there is no direct need to also provide those products with samples.. However, a sample

The present research applies the model of Vogel, Evanschitzky, and Ramaseshan (2008) in order to investigate whether the relationship between customer loyalty and its

[r]