BSc Business Administration Bachelor Thesis
___________________________________________________________________________
Spread the word - Extraversion, Gender and Online Reviews
___________________________________________________________________________ by MAÏLIS MAURICE 11594322 30th June 2020 Supervisor: Dr. D. M. DekkerFaculty of Economics and Business UNIVERSITY OF AMSTERDAM
Statement of originality
This document is written by Maïlis Maurice who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are 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.
Table of contents
Introduction………..………..…3
Theoretical framework………..…….……….……5
Methods………..……….………….12
Design, Sample and Procedure……….…..…12
Measurements……….………12
Analytical Plan……….………..…….14
Results………..……….………..……..14
Discussion and Conclusion………..………….……19
References………….………..………..…………25
Appendices….………..……….……30
List of tables Table 1: Correlation table………..……….…….………..………15
Table 2: Model summary (Hypothesis 1)……….……….…….……….…………..……16
Table 3: Regression coefficients (Hypothesis 1)…………..…….………..………..17
Table 4: Interaction effect (Hypothesis 2)….……….…..………..…..……….17
List of figures Figure 1: Conceptual model………..………12
Abstract
Online reviews have the power to make or break a business and therefore should retain managers’ attention. While a lot of previous research has discussed the importance of online reviews, the nature of the online reviewer has rarely been addressed. This study proposes that extraversion, which is characterised by sociability, talkativeness, and a greater online presence, is associated with writing online reviews. Furthermore, the study suggests that women, who engage in more communication and knowledge-sharing activities online, may influence the relationship between extraversion and online review writing. More specifically, this research hypothesises that there is a positive correlation between extraversion and online review writing (Study 1) and secondly, that this relationship is moderated by gender (Study 2). The hypotheses were tested with a convenience sample of a 140 participants. Hypothesis 1 found support, meaning that indeed, the more extraverted one is the more likely one is to write online reviews. Hypothesis 2, however, did not find support: this means that gender does not have a significant moderating effect on the relationship between extraversion and online review writing.
Spread the word - extraversion, gender and online reviews
1. Introduction
In most industries, there is an immense correlation between consumers promoting a firm via word of mouth actions, and a company’s growth rate (Reichfeld, 2003). Having this information at hand, managers should take a closer look at their consumers’ review writing practices and understand the mechanisms thereof. What exactly drives people to write reviews? Which types of customers are more likely to write reviews in the first place?
A reoccurring term along this study will be that of ‘delight’, which is when a customer has had a positive surprise going beyond his or her expectations causing a truly memorable consumer experience (Magnini, Crotts & Zehrer, 2011). Achieving this delight among your customers is of crucial importance, considering that delighted customers are more loyal to your firm which in turn increases your revenues. One study suggests that a 5% increase in customer loyalty can increase a firm’s profits from 25% to 85% (Berman, 2005). The increase in revenues is not only due to an increase in the retention rate of customers, but it is also due
to the fact that delighted customers are more likely to share their positive experience with others on online review platforms (Dekker, 2018).
Reviews can take place on well-known reviewing sites such as “Google reviews” but also as simple blog entries aimed at sharing experiences (Magnini et al., 2011). As suggested by Berman (2005), regardless of the product or experience being sold, managers and marketers should seek to text-mine these customer reviews in order to obtain better insights into the minds of customers and use those insights to better their businesses.
It is important to add that while a firm should seek out positive reviews and establish a clear online review strategy, it perhaps does not need to do so at the same intensity throughout a product’s life-cycle. This is suggested by Hu, Liu and Zhang (2008), who explain that the relationship between online reviews and sales depends on the “age” of the good, such that the longer the good exists on the market, the smaller the correlation between online reviews and sales becomes. The authors explain this phenomenon the following way: when a product has newly been introduced to a market, there is only a “limited number of sources of product quality information” and there is, therefore, a big reliance on online reviews. However, as the market gains experience with the product over time, consumers obtain information about the product through other channels and the impact of online reviews on sales goes down.
We thus understand the importance of cultivating positive reviews instead of negative reviews by delighting customers with the hopes of increasing sales and growth. What is also known, is that not everyone is equally likely to write an online review. According to PewResearch Centre (2016), 38% of American citizens claim they never leave reviews on products and services. When looking at those that do leave online reviews, people aged 50 and above are in a clear minority.
This begs the question of whether we can narrow it down even further and understand whether certain very specific personal characteristics render people more prone to writing online reviews. Perhaps certain personality traits? Maybe also personal characteristics, such as gender?
The goal of this paper will therefore be to take a closer look at these individual features and focus especially on the following two proposed attributes: level of extraversion and gender. The research question is looking at whether extraverted people, characterised by their energy,
sociability and positiveness, are more likely to write online reviews than introverted people, and whether this suggested correlation is intensified when the reviewer is female.
Gaining more knowledge in this matter is of value in the fields of behavioural science and could perhaps serve as a starting point for other, larger-scale researches. Secondly, this research is of practical relevance for managers and employees to understand differences between their customers’ inclination to write online reviews. They may be able to devote more energy and time towards delighting customers known to engage in review-writing (without neglecting other customers of course), in order to increase their company’s odds of receiving (positive) reviews. In a way, by gaining more knowledge into “who” is more likely to write reviews, companies would be able to engage in a sort of ’targeted delighting’, which could potentially maximise the number of positive reviews they receive.
The paper proceeds as follows. Section 2, the theoretical framework, will discuss all relevant literature and important findings. Section 3, the methodology, contains a precise description of the data that has been collected. Section 4 will present the results, and finally, section 5 will discuss the findings, their implications and will end with a conclusion.
2. Theoretical framework
This section will identify the key concepts of this study, define them with prominent literature, and discuss theories about their existing and proposed relationships.
There are four major concepts in this research, namely (1) online reviews, (2) customer delight, (3) extraversion, and lastly (4) gender.
Concepts can have multiple interpretations and meanings, therefore it is important to define them concisely with relevant literature.
(1) Online reviews: While informal word of mouth communication between consumers has always existed, with the advent of the Web 2.0, a new form of word of mouth communication grew exponentially, namely electronic-WOM or what is commonly called ‘online reviews’. On the Web 2.0., internet users are not simply consuming content but creating and sharing content themselves, which explains the rise and popularity of online reviews (Gonçalves, Silva & Martins, 2018).
Hennig-Thurau, Gwinner, Walsh, & Gremler (2004) define eWOM as “any positive or negative statement made by potential, actual, or former customers about a product or a company, which is made available to a multitude of people and institutions via the Internet”. This practice of writing and consuming online reviews revolutionised product and service information access, considering that before the rise of eWOM, everyone was relying on partial information, namely that provided by the company itself. Now, consumers have free access to impartial reviews, for any product category they wish and they can make better-informed purchase decisions (Couzin & Grappone, 2013).
Reviews are also not limited to one single product category or one single reviewer-category: anything can be reviewed, anyone can review, and reviewing can take place on (almost) any platform. On the internet, you will find product reviews, company reviews, reviews in the shape of social media commentary (e.g. Twitter), Peer-to-Peer Buyer and Seller ratings (Airbnb, Marktplaats…), and much more (Couzin & Grappone, 2013).
Businesses should not sleep on the impact of eWOM. For instance, in the hotel industry, positive online reviews considerably increase someone’s intention to book a hotel, whereas negative reviews reduce that intention (Ye, Law, & Gu, 2009). The link between positive (negative) online reviews and increased (decreased) sales is especially high for weak and unknown brands, as compared to more established brands that are altogether less affected by reviews. Besides the direct effect online reviews can have on increasing or decreasing sales, they also have an indirect effect on building a firm’s brand equity, i.e. the perceived worth of the brand itself (Ho-Dac, Carson, & Moore, 2013). Focusing on cultivating and promoting positive online reviews, gives managers of weaker brands a way to compete other than using traditional marketing communication strategies.
While firms should seek out positive reviews, they should not lose sight of negative reviews. Firms should make sure negative reviews remain isolated instances and not frequent occurrences, as the latter can be very damaging, especially in the hotel industry (Melián-González, Bulchand-Gidumal, & González López-Valcárcel, 2013).
It is, however, senseless to try to completely eradicate negative reviews, as that is just not feasible and some isolated cases of negative reviews may actually enhance the trustworthiness of your positive reviews. As Anderson (2014) explains it, 68% of buyers trust in reviews more
if there are also negative reviews. When your firm or your products only have positive reviews, customers might get a feeling that it is fake.
In addition, acknowledging customer complaints by replying to them is of key importance. The response should be polite, empathetic and it should seek to resolve the issue. Effective complaint management not only increases the positive feelings the initial review writer has towards the firm, but also increases positive feelings of all other individuals looking at the reviews (Sparks, So, & Bradley, 2016).
A firm should, therefore, see those “isolated cases” of negative comments as an opportunity to learn and improve the business and they should effectively communicate those intentions (Melián-González et al., 2013).
So what we have learned from online reviews is that they have become of crucial importance in the last decade with millions of people engaged in the making and consumption of eWOM. Individuals rely on reviews as much as family and friends, and positive reviews have a significant importance in boosting your sales and building your brand. While positive reviews are the best to promote your firm, a few negative reviews are inevitable and can be managed successfully to improve a firm’s overall image.
(2) Customer delight: A lot of researches have shown that the incentive for writing online reviews comes from customers having gone through either “an extremely good or extremely bad experience”. Customers that have had an average experience are significantly less likely to share electronic reviews (Melián-González et al., 2013).
It thus becomes clear that good online review management goes hand in hand with providing extremely good experiences to customers, and that managers should show more interest in going beyond simple customer satisfaction.
Extremely good customer experiences, also known as “delightful experiences”, have been discussed, measured and analysed in a growing amount of papers. Magnini et al. (2011) define ‘delight’ as “occurring when a customer receives a positive surprise beyond his or her expectations”. The reason why delight is such an important concept for firms and managers to grasp is that it highly correlates with repurchase intentions and positive WOM actions. Delight differentiates itself from simple satisfaction because the element of surprise produces unique customer emotions such as arousal, joy and pleasure, and these emotions leave a strong memory impact on the consumer.
In the hotel industry, proven ways to foster customer delight are exceptional customer service, cleanliness and hotel location (Magnini et al., 2011). In the retail industry, an employee can delight his customers if he “helps a customer see something they hadn’t considered before” like a new idea or a new product, to effectively exceed the needs of that customer (Welch & Jones, 2018).
As Dekker (2015) lays out in her article, delighted hotel guests are more likely to remain loyal and to share positive eWOM. The contrary also holds true: customers who endure the opposite of a delightful experience, namely a terrible experience, declare that they are more likely to share negative WOM about their experience (Arnold, Reynolds, Ponder & Lueg, 2005).
Positive (negative) online reviews and their positive (negative) effects on businesses caused by delightful (terrible) customer experiences hence apply to many sectors. What all managers should take from these findings is the importance of delight in keeping current customers, and in attracting new ones, namely via the spread of positive word of mouth.
In order to introduce the two following key concepts of this study, extraversion and gender, I wish to present a study led by PewResearch centre in 2016, in which American consumers were interrogated about their online review behaviour patterns. Respondents had to think of three different types of purchases: products, services and restaurant visits.
While 40% of U.S. customers revealed that they nearly always rely on online reviews for their purchase decisions, the study found that a much smaller segment of the population actually writes those online reviews. Only 1 in 10 Americans claim they almost always/always write reviews. The other 90% of Americans write reviews only occasionally (around 48%) or never do (around 41%).
This gap between consumers that (almost) always write reviews and those that never do is a very interesting topic of research. This paper seeks to understand the origin of this discrepancy between American consumers, and looks at individual personal characteristics and personality traits. Could extraverted women constitute a significant chunk of the 10% that (almost) always write online reviews?
(3) Extraversion: In line with the previously discussed paragraph, evidence has shown that the majority of internet users are passive most of the time, meaning that they do not produce their own user-generated content (UGC). Rather, they consume the content of a
minority of “active users” (Pagani, Goldsmith & Hofacker, 2013). These active consumers are spreading WOM, they are providing “customer support” to other consumers, and can build up brands. This minority of internet users thus has the power to generate value for the firm (Dwyer, 2007). For these reasons, Pagani et al. (2013) urge managers to understand who creates UGC, and why they do.
Research in the personality domain has proven that the extraversion-introversion dimension is a very important component of personality and can account for a large degree of human personality variance. Therefore it has been included in Goldberg’s Big Five inventory (1992), along with conscientiousness, agreeableness, neuroticism and openness to experience. Briefly put, extraversion is associated with energy, dominance, sociability, expressiveness and positive emotions.
The extraversion-introversion dimension is particularly interesting to measure in the context of online behaviour for several reasons. It has been proven that extraversion is positively correlated to communicative features of social networking services altogether (Correa, Hinsley, & Zuniga, 2010). Extraverts make more use of instant messaging platforms, they belong to more online groups and they have a significantly higher amount of online friends in comparison to less extraverted people (Wang, Jackson, Zhang & Su, 2012). Extraverts thus seem to have a wider audience and more people to reach with the content they create online.
The social nature of networking sites encourages extroverted people to share their thoughts and ideas with others. Extraverts were also shown to have a stronger desire to share their knowledge with others (Wang & Yang, 2007). The latter statements have been tested in a 20-country study led by Gil de Zuniga, Diehl, Huber and Liu (2017), where the results showed that, indeed, extraverted people write and post more online comments and content. Another very interesting study led by Yoo and Gretzel in 2011, shows that extraverted people feel a greater need to vent, which is the act of releasing a strong emotion. This could imply that extraverted consumers, after experiencing delightful (or terrible) customer encounters, are more likely to share their subsequent emotions.
The potential relevance to marketers is clear: extroverts not only have larger online social networks, but they are also more prone to making comments, expressing their emotions and sharing knowledge because of their sociable and talkative nature. This makes extroverts a
potentially attractive group in the promotion of a brand via word of mouth and they thus should retain the attention of managers.
These findings suggest that extraverted people, by having a bigger online presence and by producing more user-generated content, are more prone to sharing reviews. The latter assumption could be of even greater relevance if we consider review writing and the sharing of knowledge as a social activity, as extraverts are characterised by sociability and talkativeness (Wang & Yang, 2007).
Hypothesis 1 thus states that there is a positive relationship between the extraversion of the customer and his or her propensity to write an online review.
(4) Gender: Research has been looking into the different behaviour men and women exhibit online since the beginning of the internet. The general trend observed is that men are using the internet more for amusement and leisure, whereas women tend to use it for communicative and educational assistance (Jones, Johnson-Yale, Millermaier & Pérez, 2009). A major reason why women engage in more online communication is that “talk, communication and emotional expression represent the core foundation on which females build relationships” and it is thus normal for these behaviours to be reflected in the online world (Weiser, 2000).
According to Thanuskodi (2013), women spend more time on “online messaging” and “promotional campaigns” than men. Women, compared to men, are also more likely influenced by friend and purchaser recommendations than marketer information (Gretzel & Yoo, 2008). Women also tend to enjoy being part of virtual communities to give and receive support (Awad & Ragowsky, 2008). Very interestingly, a study led by Sethna, Hazari & Bergiel (2017) shows the women are significantly more influenced by online comments in their purchase decisions than men. It could thus be argued that women write more comments since they personally value online reviews when weighting their purchase decisions more than their male counterparts.
Are women more inclined to write online reviews about various customer experiences in order to help and inform other customers? Gretzel and Yoo (2008) declare that research suggests that women are more likely to engage in WOM behaviour, however, they do point out that there is a lack of research regarding demographic differences in online review writing. Wiley (2008) also states that although gender is very often analysed in the field of
consumer behaviour, it is significantly neglected in studies regarding the sharing of electronic WOM.
As suggested by the literature, women may be more inclined to write online reviews than men. This paper seeks to take this proposed relationship and add subtlety to it, namely that the female gender would act as a moderator in the previously proposed relationship between extraversion and eWOM. In simple words, this paper will test whether extraverted women write more online reviews than extraverted men.
Hypothesis 2 thus states that there is an interaction effect between extraversion of the customer and gender on the propensity to write an online review, such that the relationship between extraversion and the propensity to write an online review becomes stronger when a customer is female rather than male.
An additional reason for why it is interesting to focus on women rather than men, is that just like extraverted people, they have more friends on social networking services and can thus potentially reach a wider audience with the content they create and the reviews they write (Wang, Jackson, Zhang & Su, 2012). Furthermore, gender is an important variable in marketing and is often used in a practical way to segment customers, hence why it is compelling to look at female-male differences in the domain of eWOM (Yelkur & Chakrabarty, 2006).
The four main concepts of this study have now been clarified, and their relationships explained.
Online reviews (1) have the power to make or break a business, therefore managers should have a proactive online review strategy where positive reviews are encouraged. One way to do so is to delight (2) your customers. That is to provide your customers with exceptional and memorable experiences which will highly increase their likelihood of spreading positive word of mouth about your firm, in real life and online. Extraversion, (3) someone’s degree of sociability and talkativeness, is associated with a greater output of user-generated content, such as online reviews. Finally, the influence of gender (4) on online behaviour and the likelihood of writing online comments was assessed: research suggests that women are more prone to writing online reviews because of their informative and communicative online nature.
Figure 1. Conceptual model
3. Methods
Design, Sample and Procedure
In order to test these two hypotheses a cross-sectional design was best fitted. The data was collected using a survey questionnaire of 45 questions.
The type of sampling used was convenience sampling, which is a part of non-probability sampling. Two other students and I recruited participants using our personal contacts and various online groups at hand. Participation in the study was on a voluntary basis, without monetary or other compensation. Considering our cooperation, the distributed survey evaluated more variables than are going to be examined in this research paper.
We collected 171 responses in 2 weeks, between April and May 2020. Out of the 171 responses, 21 responses were empty and 10 were incomplete. The response rate is therefore of 82%. The empty responses and the partially completed responses were deleted as they would have negatively influenced the accuracy of the results of this study. The study thus has a sample size of N = 140.
Women were in a clear demographic majority, making up 62.9% of the sample’s respondents. The ages of respondents varied between 14 and 70 years old (M= 29.5, SD = 13.75). Furthermore, the majority of respondents were from the Netherlands (62,1%), followed by Luxembourg (14.3%) and France (5%).
Measurements
This study focuses on 3 variables: (1) the propensity to write online reviews, the outcome variable, (2) extraversion, the predictor variable, and (3) gender, which is the moderator variable. All three variables were rated by the same respondents. Most questions were
recorded on 5- and 7-point Likert scales. For certain inquiries, such as gender and age, multiple-choice and open-ended questions were the most appropriate.
(1) Propensity to write online reviews was measured by interrogating respondents about their past and their general/hypothetical review writing. 11 items were specifically created for the purpose of measuring online reviewing. Past review writing behaviour was simply measured with the following item “How many online reviews have you approximately written in the past year?”. General/Hypothetical review writing was measured with 10 different items that were grouped into two different components with the help of a factor analysis. The first component, named ‘negative online reviewing’, contained 5 items such as “In general I tend to write negative reviews” or “How likely are you in this scenario to leave a negative online review?” which was a follow-up question to a negative customer experience scenario. This scale had a Cronbach’s alpha of .807, which is good. The
second scale was named ‘positive online reviewing’ and contained 5 items such as
“Please indicate how likely you are to write an online review after a delightful experience” or “How likely are you in this scenario to leave a positive online review?”. The latter item is, again, a follow-up question to a customer experience scenario, namely the following one: “Imagine that you are ordering a pair of earrings online from a brand you trust a lot, where you have ordered 7 times before and you always got the best value for your money. Three days later, you receive a package bigger than what you would have expected. Intrigued, you open it and realise that the company added a purse you did not order and a letter saying "here is a little thank you for being such a loyal customer at RozeJewellery”. This scale’s Cronbach’s alpha was of .850 which is a good score. Hypothesis 1 and 2 will be tested with a variable named ‘online review writing’ that was created by combining all 10 items pertaining to positive and negative online review writing. This 10-item scale had a Cronbach’s alpha of .850 and is therefore good as well. (2) Extraversion was measured using the revised NEO personality scale developed by Costa
and McCrae in 2008. The scale contains 12 items concerning extraversion. Example items are ‘I really enjoy talking to people’, ‘I am a very active person’ or ‘I usually prefer to do things alone’, the latter being one of the 4 reverse-scaled items that needed to be recoded. The extraversion scale’s Cronbach’s alpha was of .777, which is reasonable.
(3) Gender was measured by asking the question “What is your gender?”. There were three possible responses: ‘male’, ‘female’, and ‘other’.
(4) Lastly, control variables were added to the study so as to cancel out alternative effects on the hypotheses. The chosen control variables were age and the level of conscientiousness. Conscientiousness was chosen as a control variable considering that Wang and Yang (2007) wrote that “extraversion, agreeableness and conscientiousness are positively related to an individual’s intention to share knowledge”. Furthermore, people scoring high on conscientiousness are more likely to be thoughtful and mindful of others (Morrison, Cheong & McMillan, 2013). This leads to think that conscientious people may engage in more review writing so as to help or warn others in their decision making. Conscientiousness was measured using the NEO personality scale by Costa and McCrae (2008), just like for extraversion. The scale regarding conscientiousness also contained 12 items. Examples of those are “I strive for excellence in everything I do” and ‘I keep my belongings neat and clean’. In this scale as well, 4 items were reverse scaled, which therefore needed to be recoded. The conscientiousness’ scale had a Cronbach’s alpha of .680, which is a reasonable score. Age was assessed by asking ‘What is your age?’, to which respondents simply filled out their age in an empty box
Analytical plan
In order to test Hypothesis 1, the positive relationship between extraversion and the likelihood of writing online reviews, a linear regression will be applied with ‘extraversion’ as the predictor variable and ‘online review writing’ as the outcome variable.
Hypothesis 2, the interaction effect between extraversion and gender on the likelihood of writing an online review, will be tested using PROCESS macro of Hayes (2018). ‘Extraversion’ will be the predictor variable, ‘gender’ will be the moderating variable and ‘propensity to write an online review’ will be the outcome variable.
4. Results
A correlation table was created to get some initial insights into the data. Table 1 contains the means, standard deviations and correlations of the main construct variables and control variables, except for gender. To get an idea whether gender correlates with the construct and
control variables, an independent samples t-test was conducted. However, this preliminary test showed no significant effect of gender on any of the constructs (see Appendix B).
TABLE 1.
Descriptive statistics and correlations
*Note *p<.05, **p<.01, ***p<.001
The results displayed in table 1. are very interesting. Hypothesis 1 stated that an individual’s
extraversion level positively correlates with writing online reviews. The correlation table
indicates that extraversion is indeed strongly correlated with the amount of reviews written last year (r = .27**; p = .001), positive review writing (r = .25**; p = .003), and online review writing as a whole (r = .24**; p = .005). The correlations being of .27, .25 and .24, they can be qualified as moderate, with very significant p-values. Extraversion also has a significant correlation with negative review writing (r = .15* ; p = .04). However, that correlation is weaker, being of .15 and having a p-value greater than .01. It is nevertheless an interesting correlation that will be looked at further in the last part of the results section, namely in contrast with positive review writing.
These results could signal that, as H1 predicted, extraverted people write more online reviews than non-extraverts.
Unexpectedly, conscientiousness has a significant negative correlation with the amounts of reviews written in the past year (r = -.14* ; p = .04). This could imply that conscientious people are not more likely to write online reviews, which would contradict what had previously been proposed.
Variables M SD 1. 2. 3. 4. 5. 6. 7.
1. Online Review Writing 22.5
6
1.16
-2. Positive Online Review Writing 23.3 2
1.40 .85**
-3. Negative Online Review Writing 21.8 0
1.34 .84** .43**
-4. Amount of reviews written in the past year 1.92 1.31 .45** .48** .27** -5. Age 29.5 0 13.75 .00 .08 -.08 -.06 -6. Conscientiousness 3.46 .50 -.10 -.07 -.10 -.14* -.03 -7. Extraversion 3.56 .53 .24** .25** .15* .27** -.07 .11
-Lastly, age has no significant effect on any of the study’s variables.
Before any testing could be done, assumptions needed to be verified so as to make sure no false conclusions would later be drawn from the data analysis (see appendix C). The first assumption, which states that there must be a linear relationship (1) between extraversion and online review writing, was met. The second assumption concerned the
normality of residuals (2) and was also met. Then, the homoscedasticity of residuals (3) was
confirmed, as the residuals were equally variable. There was also no multicollinearity (4) since all the independent variables had VIF-scores largely inferior to 5. Finally, the fifth assumption concerned outliers (5): Cook’s distance values were all below 1 which means that there were no significant outliers. Considering all assumptions were met, the data can be manipulated and interpreted with confidence.
In order to test hypothesis 1, which stipulates that there is a positive relationship
between extraversion and online review writing, a simple hierarchical regression test was run.
In a first instance the test contained control variables only (model 1) and in a second instance, the predictor variable ‘extraversion’ was added (model 2), so as to see if the latter model enhanced explanatory power in the dependent variable ‘online review writing’. The results were the following:
TABLE 2. (Study 1) Model summaryc
a. Predictors: (Constant), Age, Conscientiousness
b. Predictors: (Constant), Age, Conscientiousness, Extraversion c. Dependent variable: Online review writing
As can be drawn from the table, model 2 has an R-square of .073: This means that 7.3% of the variance in online review writing can be explained by model 2, whereas only 1% of the variance is explained by model 1. Model 2 thus explains for more variance in the dependent variable ‘online review writing’ than model 1.
Model R R-square Adjusted
R-square Std. Error of the Estimate R Square change F Change df1 df2 Sig. F change
1 .101a .010 -.004 1.16 .010 .709 2 137 .494
R-square went up by .063 when including extraversion (model 2) with a p-value of .003 (p < 0.005). This indicates that the change of explained variance caused by adding extraversion to the model was significant and it thus confirms that model 2 is more adequate than model 1 in predicting online reviewing.
TABLE 3. (Study 1)
Regression coefficients results for the 2 models
Finally the beta of the predictor variable, extraversion, is equal to .581 with a t-value of 3.028 and a p-value of .003 (p < .005). The standard error is equal to .192 and has a 95% confidence of [.201, .960] (see appendix D). This implies that the predictor variable ‘extraversion’ is a significant predictor of the outcome variable. The unstandardised beta coefficient reveals that when extraversion is augmented by ‘one unit’, online review writing is augmented by .581, assuming all other variables remain constant.
Taken together, there is enough statistical evidence in support of hypothesis 1 and it can thus be stated that there is a positive relationship between extraversion and online review writing.
To test hypothesis 2, which states that there is an interaction effect between
extraversion and gender on online review writing, such that the relationship between extraversion and online review writing becomes stronger when the reviewer is female,
PROCESS macro of Hayes (2018) was used. TABLE 4. (Study 2)
Results for the interaction effect between extraversion and gender on online review writing
Model Unstandardised Coefficients
Beta | SE Standardised Coefficients Beta t Sig. 1. (Constant) Conscientiousness Age 23.355 | .707 -.221 | .185 .000 | .007 -.101 -.003 33.038 -1.191 -.031 .000 .236 .975 2. (Constant) Conscientiousness Age Extraversion 21.507 | .919 -. 278 | .181 .001 | .007 .581 | .192 -.128 .014 .252 23.404 -1.536 .166 3.028 .000 .127 .868 .003 b se t p Constant 23.59 .76 30.87 .000
The variable gender on its own is largely insignificant on online review writing (p = .716 > 0.05), which reaffirms the results from the t-test that was carried out earlier.
The interaction term between extraversion and gender, accounts for 1,03% additional variation as is indicated by the R-square value (see appendix E). However, this interaction, by having a p-value superior to the cut-off value of .05, is insignificant (b = -.42, se = .31, t = -1.34, p = .181 95%, CI = [-1.04, .20]).
It can therefore be concluded that there is not enough evidence in support of
hypothesis 2, and that gender therefore does not have a statistically significant interaction
effect with extraversion on online review writing.
The last part of the results section concerns the difference between positive and negative online review writing that was hinted at by the correlation table: extraversion seemed to correlate more significantly with positive reviewing rather than with negative reviewing. A linear graph was therefore created in an effort to illustrate this nuance, by sketching each relationship’s function. Furthermore, two linear regressions were conducted to compare the results meaningfully. All assumptions have been tested and met (see appendix C).
FIGURE 2. (Sudy 3)
Extraversion as a predictor of Positive and Negative Online Reviewing
Extraversion (x) .59 .18 3.30 .001 Gender (w) .07 .20 .36 .716 Age .00 .00 .09 .930 Conscientiousness -.29 .22 -1.33 .185 X*W -.42 .31 -1.34 .181 b se t p 0 5 10 15 20 25 30 35 0 2 4 6 8 10 12 14 16 Na tu re o f On li ne R ev ie wi ng Extraversion
Postive and negative online reviewing compared
Online Reviewing
Postive Online Reviewing
As figure 2 shows, extraversion is positively correlated with all three types of online reviewing: combined positive and negative online reviewing, positive online reviewing and negative online reviewing. However, extraversion is shown to correlate the strongest with positive online review writing, followed by online review writing and lastly negative online review writing.
Note that it is logical for online review writing to “fall in the middle” as it is composed of both positive and negative online review writing.
The results of the two simple linear regressions confirm these observed trends (see appendices F and G): When adding extraversion to the model pertaining to positive online reviewing, r-square changed by .070 whereas when extraversion was added to the model pertaining to negative reviewing, r-square changed by .024 only. This means that extraversion explains the variance in the outcome variable ‘positive online reviewing’ to a greater extent than it explains the variance in the outcome variable ‘negative online reviewing’.
The p-values were also telling: model 2 had a p-value of .002 < .005 when the dependent variable was positive reviewing, but had a p-value of .068 > .05 when the dependent variable was negative online review writing. This means that the increased variance explained by extraversion on positive online review writing was significant, whereas the increase in variance explained by extraversion on negative review writing was not significant. The latter result is surprising as the the correlation table showed a significant correlation between extraversion and negative online reviewing.
To conclude, extraversion has as significant linear correlation with ‘positive online reviewing’ but not with ‘negative online reviewing’.
5. Discussion
Findings
This study aimed to investigate whether extraversion positively correlated with the propensity to write online reviews and whether this proposed relationship was moderated by gender.
The results lend support for the first hypothesis, which stated that there is a positive
relationship between extraversion and online review writing. This signifies that extraverted
The results did not, however, provide support for the second hypothesis, which stated that
there is an interaction effect between extraversion and gender on online review writing, such that the relationship between extraversion and online review writing becomes stronger when the reviewer is female. This means that a woman scoring high on extraversion is not more
likely to engage in online review writing than a man scoring high on extraversion.
Additionally, it was found that extraversion has a significant linear correlation with positive online reviewing but not with negative online reviewing. This means that a person scoring high on extraversion is more likely to write positive rather than negative online reviews.
Interpretation of results and alternative explanations
The results relating to hypothesis 1 are in line with what had been predicted in the theoretical framework and they add novelty to previous research. Numerous papers have addressed the relationship between extraversion and user-generated content. Pagani et al. (2013) for instance stated that people scoring high on extraversion make greater use of Facebook communication components than people scoring lower on extraversion. Extraverted people have a longing for expressing themselves and interacting socially online. Online review writing can be defined both as a “social interaction” and it falls in the category of “user-generated content” (Goswami, Park & Song, 2017). Considering that extraverted people are proven to seek out social interactions and to produce more online content, the finding that extraverted people write more online reviews than non-extraverts is thus very much in keeping with previous literature. Lastly, studies have also shown that extraversion is positively correlated with assertiveness, which is embedded in online review writing (Ames & Flynn, 2007).
Having said that, the main difference between previous papers and this study is the scope of the research. While most previous research papers have documented the relationship between extraversion and ‘online social interactions’ or ‘user-generated content’ altogether, this study took a more micro-approach by analysing one of the many forms of user-generated content on its own, that is online reviews. This research thus aimed to account for more nuances within previous research, and with the use of a simple regression analysis, showed that the more extraverted one is, the more likely one is to write online reviews.
Testing for hypothesis 2 showed that gender has no impact on the relationship between extraversion and online review writing. This unexpected result is very likely due to the fact that the proposed theory was flawed to begin with due to a lack of preexisting theory. Very few articles have been published in the domain of online reviewer demographics and even lesser in the domain of gender and online reviews specifically. Most articles found were addressing general female vs. male online behaviour patterns and were not specific to online reviewing. In fact, as mentioned earlier, Gretzel and Yoo (2008) emphasised in their study that there is a major lack of research regarding demographic differences in online review writing. Wiley (2008) also affirmed the latter statement.
Furthermore, the majority of literature discusses differences in male-female online behaviour in the 2000’s. These online gender patterns may thus be outdated considering that in the past 10 years internet use has rapidly changed and internet access has immensely increased globally (Kemp, 2019).
The limited number of academic papers that do address women’s online behaviour state that women exhibit more communication and knowledge-sharing behaviour, whereas men rather use the internet to surf online (Postmes & Spears, 2002; Jones, Johnson-Yale, Millermaier & Pérez, 2009). Also, a study led by Statista in 2017 showed that 77% of female online shoppers had left online reviews in the past year, whereas for men that number was of 64%. However, these findings need to be taken with a grain of salt, as Weiser (2000) explains that gender differences on the internet may appear in very specific settings only and that they depend upon age and individual internet experience.
In sum, the scarce amount of preexisting research and its possible outdatedness led to the inductive inferring of hypothesis 2, which turned out not to be supported by the results of PROCESS macro of Hayes (2018): gender does not not have an interaction effect in the relationship between extraversion and online review writing.
Finally, the strong relationship observed between extraversion and positive review writing, contrasted with the weaker and insignificant relationship with negative review writing, is also aligned to common theories surrounding extraversion. Extraversion has been proven to be strongly associated with happiness and optimism, which are qualities that are reflected in the writing of positive online reviews (Haddock & Rutkowski, 2014). Extraverts, on top of being rather positive, are talkative, social and assertive individuals (Hermes,
Hagemann, Neumann & Walter, 2011). The finding that extraverts are more inclined to write positive online reviews is therefore in harmony with the very cornerstone of extraversion, that is engaging in social interactions (online reviewing) positively and assertively.
Limitations
One of the biggest drawbacks of this study lies in the methodology. Even though a sample of 140 is sufficient, a bigger sample is always preferred and would have added more statistical power to the study. This could have perhaps led to different conclusions regarding the hypotheses.
In addition, it is critical to realise that the type of sampling used, convenience sampling, may hinder the generalisability of this study’s results since a majority of respondents were around the same age, from the same region (Benelux) and had similar life occupations. Along the same lines, 62.9% of the respondents were female and males were therefore under-represented. It can be argued that if a simple random sampling method had been used, which would have led to a more diverse sample, more representative outcomes would have been yielded.
Finally, another possible weakness is that the ‘online review writing’ scale was not a pre-existing scale with renowned credibility, as it was created for the purpose of this study. This may mean that the variable online review writing was not measured in the most precise and efficient way possible.
It is important to note that these presented limitations represent minor flaws in the study and do not discredit the research conducted: they simply may have had an impact on the results, which is necessary to be aware of.
Future research
Considering that this study has unanticipatedly shown that extraversion is strongly correlated with positive online reviewing but not with negative online reviewing, it may be interesting to look into this relationship further by adding mediation and/or moderator variables. Perhaps adding mediation/moderator variables could explain contextual differences that drive extraverts to write positive reviews rather than negative reviews.
Future research could also look closely to both sides of the variable extraversion, that is extraversion and introversion. That way, inferences could also be made about introverted individuals and their online review writing patterns.
Additionally, as follows from the limitations section, it would be be judicious to increase the sample size and to, as far as practicable, make use of a probability-sampling method to increase the study’s generalisability. On top of this, it would be interesting for future research to construct the same research in a different location with a different culture, in order to see if the findings remain true.
Lastly, it would be insightful to conduct new research in the domain of online reviewer demographics and online gender-differences to update previous, outdated research.
Practical implications
A very interesting article written by Matz, Kosinski, Nave and Stillwell (2017), explains that psychological targeting is a fruitful approach in persuading people online. They write: “Recent research shows that people’s psychological characteristics can be accurately predicted from their digital footprints, such as their Facebook Likes or Tweets”. In turn, individuals can then be targeted with personality-matching content. Their study found that people who were matched with their extraversion and openness-to-experience scores were 40% more likely to click on the ad and 50% more likely to make purchases than individuals that have not received personality-targeted ads.
Knowing that big data is growing in its ability to successfully identify people’s personality traits based on their online footprints, perhaps firms could make use of this phenomenon to pursue their online review strategies?
Managers of firms with online presences, from e-boutiques to firms offering online payment solutions, could screen potential and existing clients’ extraversion levels using these big data algorithms. Being aware of who their extraverted clients are, they could then make it their mission to delight them instead of just satisfying them, so as to foster the spread of eWOM and traditional word of mouth (Dekker, 2018). This would not imply that firms would neglect their non-extraverted consumers, but simply that they would spend their energy most efficiently and profitably according to which customer they are serving: the aim would be for
every customer to get at least highly satisfactory customer experiences, with a focus on delivering delightful experiences to extraverted individuals.
Furthermore, considering that extraverted people tend to write positive reviews over negative reviews, this strategy could show to be even more profitable for firms, as we have seen that positive online reviews and increased sales are very strongly correlated (Hu et al. 2008). It is important to remember that, as was highlighted in the theoretical framework, some negative reviews are normal and should also be welcomed by managers as they increase a firm’s credibility (Anderson, 2014).
Evidently, an Artificial Intelligence tool that scans an individual’s personal characteristics, and the proposed scenario of firms using this tool to their own benefits, may encounter obstacles with regards to legislation that seeks to control consumer-data exploitation and that protects individual privacy in the digital age (Matz et. al., 2017). For instance, all EU companies must comply with the General Data Protection Regulation at all times (Wolford, 2019).
The practical implications just presented are therefore still tentative in nature and are not currently in action, as they first need to be weighed out by legislative bodies. Their relevance in business practices does, however, merit to be considered and discussed.
In hospitality, where positive online reviews are very much linked to increased reservations, employees may be able to get a feel about a guest’s extraversion level and accordingly provide them with memorable experiences that spur positive word-of-mouth actions. However, as extraversion remains a personality trait and not a visible characteristic such as gender, wrong inferences may be made as situational factors such as being under the influence of alcohol, can influence someone’s level of extraversion.
While it is of course advantageous for firms with proactive online review strategies to use their energy and resources the most cost-efficient way possible by engaging in ‘targeted-delighting’, if that shows not to be (yet) possible, a company’s best bet is simply to delight all customers. That way, all odds are in the company’s favour.
Conclusion
Previous research has addressed the importance of online review writing and its link with delight and sales, without however distinctly addressing whether certain personality traits and individual characteristics may influence the act of writing online reviews. That is
why this study proposed to examine whether specific traits influence online review writing: it looked at whether there was a direct relationship between extraversion and online review writing, and whether gender acted as a moderator on the former relationship. Results showed that indeed, extraversion significantly influenced someone’s likelihood of writing online reviews. The results also showed that gender did not have a moderation effect on the relationship between extraversion and online review writing. Furthermore, an unexpected subtlety came to light, namely that extraverted people are more likely to write positive compared to negative reviews.
This study not only deepens existing theory by showing that a personality trait such as extraversion is a significant predictor of online review writing, but it also serves as a basis for a more targeted and personal way of inciting customers to write online reviews about a firm and its offerings, for instance by engaging in targeted delighting.
References
Ames, D. R., & Flynn, F. J. (2007). What breaks a leader: the curvilinear relation between assertiveness and leadership. Journal of personality and social psychology, 92(2), 307.
Anderson, M., & Anderson, M. (2014, July). 88% of consumers trust online reviews as much
as personal recommendations. Retrieved from https://searchengineland.com
Ariffin, A. A. M., & Maghzi, A. (2012). A preliminary study on customer expectations of hotel hospitality: Influences of personal and hotel factors. International Journal of
HospitalityManagement, 31(1), 191-198.
Arnold, M. J., Reynolds, K. E., Ponder, N., & Lueg, J. E. (2005). Customer delight in a retail context: investigating delightful and terrible shopping experiences. Journal of
Business Research, 58(8), 1132-1145.
Berman, B. (2005). How to delight your customers. California Management Review, 48(1), 129-151.
Bitner, M. J., Booms, B. H., & Mohr, L. A. (1994). Critical service encounters: The employee’s viewpoint. Journal of marketing, 58(4), 95-106.
Clement, J. (2018, June 26). U.S. online shoppers online reviews by gender 2017. Retrieved June 19, 2020, from https://statista.com/statistics/184037
Correa, T., Hinsley, A. W., & De Zuniga, H. G. (2010). Who interacts on the Web?: The intersection of users’ personality and social media use. Computers in human
behavior, 26(2), 247-253.
Costa Jr, P. T., & McCrae, R. R. (2008). The Revised NEO Personality Inventory (NEO-PIR). Sage Publications, Inc.
Couzin, G., & Grappone, J. (2013). Five stars: putting online reviews to work for your business. John Wiley & Sons.
Dekker, D. M. (2018). Genuinely hospitable behaviour in education. In Innovation in
Hospitality Education (pp. 65-75). Springer, Cham.
Dwyer, P. (2007). Measuring the value of electronic word of mouth and its impact in consumer communities. Journal of Interactive marketing, 21(2), 63-79.
Gil de Zúñiga, H., Diehl, T., Huber, B., & Liu, J. (2017). Personality traits and social media use in 20 countries: How personality relates to frequency of social media use, social media news use, and social media use for social interaction. Cyberpsychology,
Behaviour, and Social Networking, 20(9), 540-552.
Goldberg, L. R. (1992). The development of markers for the Big-Five factor structure.
Psychological assessment, 4(1), 26.
Gonçalves, H. M., Silva, G. M., & Martins, T. G. (2018). Motivations for posting online reviews in the hotel industry. Psychology & Marketing, 35(11), 807-817.
consumer review fraud detection. Journal of Big Data, 4(1), 1-19.
Gretzel, U., & Yoo, K. H. (2008). Use and impact of online travel reviews. Information and
communication technologies in tourism 2008, 35-46.
Haddock, A., & Rutkowski, A. (2014). Psychology of extraversion . Nova Science Publishers, Incorporation.
Hennig-Thurau, T., Gwinner, K. P., Walsh, G., & Gremler, D. D. (2004). Electronic word-of- mouth via consumer-opinion platforms: what motivates consumers to articulate themselves on the internet?. Journal of interactive marketing, 18(1), 38-52.
Hermes, M., Hagemann, D., Naumann, E., & Walter, C. (2011). Extraversion and its positive emotional core—Further evidence from neuroscience. Emotion, 11(2), 367.
Ho-Dac, N. N., Carson, S. J., & Moore, W. L. (2013). The effects of positive and negative online customer reviews: do brand strength and category maturity matter?. Journal of
Marketing, 77(6), 37-53.
Jones, S., Johnson-Yale, C., Millermaier, S., & Pérez, F. S. (2009). US college students’ Internet use: Race, gender and digital divides. Journal of Computer-Mediated
Communication, 14(2), 244-264.
Kemp, S. (2019, January 30). Digital 2019: Global Digital Overview - DataReportal – Global Digital Insights. Retrieved from https://datareportal.com/reports/digital-2019-global-
Magnini, V. P., Crotts, J. C., & Zehrer, A. (2011). Understanding customer delight: An application of travel blog analysis. Journal of Travel Research, 50(5), 535-545.
Matz, S. C., Kosinski, M., Nave, G., & Stillwell, D. J. (2017). Psychological targeting as an effective approach to digital mass persuasion. Proceedings of the national academy of
Melián-González, S., Bulchand-Gidumal, J., & González López-Valcárcel, B. (2013). Online customer reviews of hotels: As participation increases, better evaluation is
obtained. Cornell Hospitality Quarterly, 54(3), 274-283.
Morrison, M. A., Cheong, H. J., & McMillan, S. J. (2013). Posting, lurking, and networking: Behaviors and characteristics of consumers in the context of user-generated
content. Journal of Interactive Advertising, 13(2), 97-108.
Reichheld, F. F. (2003). The one number you need to grow. Harvard business review, 81(12), 46-55.
O’connor, P. (2008, January). User-generated content and travel: A case study on Tripadvisor. com. In ENTER (Vol. 2008, pp. 47-58).
Pagani, M., Goldsmith, R. E., & Hofacker, C. F. (2013). Extraversion as a stimulus for user- generated content. Journal of Research in Interactive Marketing.
Postmes, T., & Spears, R. (2002). Behavior online: Does anonymous computer communication reduce gender inequality?. Personality and Social Psychology
Bulletin, 28(8), 1073-1083.
Sethna, B. N., Hazari, S., & Bergiel, B. (2017). Influence of user generated content in online shopping: impact of gender on purchase behaviour, trust, and intention to
purchase. International Journal of Electronic Marketing and Retailing, 8(4), 344-371.
Smith, A., & Anderson, M. (2016, December). Online reviews and ratings. Retrieved June 19, 2020, from https://www.pewresearch.org/internet/2016/12/19/online-reviews/
Sparks, B. A., So, K. K. F., & Bradley, G. L. (2016). Responding to negative online reviews: The effects of hotel responses on customer inferences of trust and concern. Tourism
Thanuskodi, S. (2013). Gender differences in internet usage among college students: A comparative study. Library Philosophy and Practice (e-journal). Paper 1052.
Wang, C. C., & Yang, Y. J. (2007). Personality and intention to share knowledge: An
empirical study of scientists in an R&D laboratory. Social Behavior and Personality:
an international journal, 35(10), 1427-1436.
Wang, J., Jackson, L., Zhang, D., & Su, Z. (2012). The relationships among the Big Five Personality factors, self-esteem, narcissism, and sensation-seeking to Chinese University students’ uses of social networking sites (SNSs). Computers in Human
Behavior, 28(6), 2313-2319.
Weiser, E. B. (2000). Gender differences in Internet use patterns and Internet application preferences: A two-sample comparison. Cyberpsychology and behavior, 3(2), 167-178.
Welch, A., & Jones, J. (2018, January). A new purpose for in-store retail in 2018: Creating moments of customer delight. HMC Sales, Marketing and Alliances Excellence
Essentials, Retrieved from https://search.proquest.com/docview/2043702682?
Wolford, B. (2019, February 13). Does the GDPR apply to companies outside of the EU? Retrieved June 25, 2020, from https://gdpr.eu/companies-outside-of-europe/
Ye, Q., Law, R., & Gu, B. (2009). The impact of online user reviews on hotel room sales.
International Journal of Hospitality Management, 28(1), 180-182.
Yelkur, R., Chakrabarty, S., 2006. Gender differences in service quality expectations in the fast food Industry. Services Marketing Quarterly 27 (4), 141–151.
Appendices
Appendix A: Survey Items
i. Variable measured: Extraversion
Construct Scale items (5-point likert scale) Extraversion 1. I like to have a lot of people around me
2. I laugh easily
3. I don’t consider myself “light-hearted” 4. I really enjoy talking to people 5. I like to be where the action is 6. I usually prefer to do things alone 7. I often feel as if I’m bursting with energy 8. I am a cheerful, high-spirited person 9. I am not an optimist
10. My life is fast-paced 11. I am a very active person
12. I would rather go my own way than be a leader of others
ii. Variable measured: Conscientiousness
Construct Scale items (5-point likert scale)
Conscientiousness 1. I keep my belongings neat and clean
2. I’m pretty good about pacing myself so as to get things done on time 3. I am not a very methodical person
4. I try to perform all the tasks assigned to me conscientiously
5. I have a clear set of goals and work towards them in an orderly fashion 6. I waste a lot of time before settling down to work
7. I work hard to accomplish my goals
8. When I make a commitment, I can always be counted on to follow through 9. Sometimes I’m not as dependable or reliable as I should be
10. I am a productive person who always gets the job done 11. I never seem to be able to get organised
Appedix B: Correlation Table & Independent Samples T-test
i. Correlation table: Means, Standard Deviations, and Correlations
iii. Variables measured: Past review writing and positive/negative review writing-1
Construct Scale items (7-point likert scale) Past online
review behaviour 1. How many reviews have you approximately written in the past year? Positive online
review writing 1. In general, I tend to write positive reviews.
2. Please indicate how likely you are to write an online review... - After a positive customer experience 3. Please indicate how likely you are to write an online review... - After a delightful experience
4. How likely are you in this scenario to leave an online review? (positive customer experience scenario) 5. How likely are you in this scenario to leave an online review? (positive customer experience scenario) Negative online
review writing 1. In general, I tend to write negative reviews.
2. Please indicate how likely you are to write an online review... - After a negative customer experience 3. Please indicate how likely you are to write an online review... - After a terrible experience
4. How likely are you in this scenario to leave an online review? (negative customer experience scenario) 5. How likely are you in this scenario to leave an online review? (negative customer experience scenario)
ii. Independent-Samples T-test with gender as the grouping variable
Appendix C: Assumption testing for regression analysis
Assumption 1: Linear relationship
Linear relationship between extraversion and positive online review writing
Assumption 2: Normality of residuals - P-plot
Appendix D: Simple regression analysis - Online Review Writing
Model summary
Coefficients table
Appendix E: PROCESS macro of Hayes (2018)
APPENDIX F: Simple Regression Analysis - Positive Online Review Writing
Model summary
ANOVA table
APPENDIX G: Simple Regression Analysis - Negative Online Review Writing
Model summary
