TO GOSSIP OR NOT TO GOSSIP: REACTIONS TO A PERCEIVED REQUEST TO GOSSIP – A QUALITATIVE STUDY

Abstract

This research primarily aims to evaluate traits exhibited by people when asked to describe in writing a third person to a friend. These traits were classified as positive or negative based on a qualitative analysis conducted on 2230 respondent texts. The request was, in fact, perceived by many participants as a request to gossip, and was opposed with various tacit and explicit rejections. Traits were analyzed and rated by software that was developed specifically for this research. The majority of traits were found to be positive, with 'good' the most prominent. This analysis also confirmed that people do not want to be perceived as gossips. They tend to oppose the request to gossip and prefer to use positive trait descriptors.

Headnote

Abstract. This research primarily aims to evaluate traits exhibited by people when asked to describe in writing a third person to a friend. These traits were classified as positive or negative based on a qualitative analysis conducted on 2230 respondent texts. The request was, in fact, perceived by many participants as a request to gossip, and was opposed with various tacit and explicit rejections. Traits were analyzed and rated by software that was developed specifically for this research. The majority of traits were found to be positive, with 'good' the most prominent. This analysis also confirmed that people do not want to be perceived as gossips. They tend to oppose the request to gossip and prefer to use positive trait descriptors.

Keywords: gossip, traits, qualitative, person descriptions

1. Introduction

Gossip is a behavior familiar to everyone. However, no one really wants to be perceived as a gossip regardless of its many advantages. According to Harari (2014), a historian who studies the origin of mankind, gossip began to fill a social role thousands of years ago and it had a significant impact on the transformation of Homo sapiens into the planet's dominant species. He claims that one of the evolutionary tools that enabled humans to organize themselves into larger groups is the ability to communicate and, more specifically, to gossip. Gossip facilitates monitoring community member behavior and interrelationships. Anthropologist and evolutionary psychologist, Robin Dunbar, has theorized that gossip is an efficient tool for forming alliances, although usually at the expense of the gossip target (Dunbar, 2008). Studies have enumerated many current advantages of gossip (for example, DiFonzo and Bordia 2007). Nevertheless, historically gossip has a bad reputation (Peters and Kashima 2015). This may be why such a common phenomenon has been studied relatively less than other behaviors (Michelson et al. 2010). Cole and Scrivener (2013) referred to this research gap, concluding that research on the causes and consequences of certain types of gossip behavior is scarce, and there are still many unanswered questions on the mechanics of gossip and its long and short term behavioral effects.

This research employed qualitative methods and attempts to fill the explanatory gap on gossip. Its aim is to understand what people think about gossiping by assessing their reactions to what they perceived as being asked to gossip. The content of their text responses was then subjected to analysis. Qualitative methods were used so as to expose hidden layers of meaning and provide deeper analysis (Ben-Hador 2016a) in order to understand better the gossip mechanism.

2.Literature review

2.1.Gossip

Gossip can be defined as an informal communication process that includes information about the social environment or other people (Noon and Delbridge 1993). Generally, gossip concerns absent third parties (McDonald et al. 2007). However, gossip is not simply a mode of information transference. In fact, Robinson (2016) suggested that gossip is generally encoded with some salacious or amusing item of information about a third party, although Cuonzo (2008) claims that any utterance can be considered gossip under certain conditions and not gossip under others.

Research has shown that gossip performs many functions in everyday life, such as channeling information and clarifying ambiguity (DiFonzo and Bordia 2007), as well as establishing certainty (Rosnow 1988). Waddington and Fletcher (2005) found that gossip is used to express a wide range of positive and negative emotions.

Gossip plays an important role in social groups (Ben-Hador, 2016b), with the capacity to create social comparison and define social limits (Hafen 2004). It can also preserve and enforce group values and norms (Grosser et al. 2012) and distinguish negative influence for group protection (Beersma and Van Kleef 2012).

Gossip can garner advantages for the gossiper (Einat and Chen 2012), supplying pleasure and entertainment, strengthening social ties (Michelson and Mouly 2000) and support (Chua and De la Cerna 2014), and displaying influence (Grosser et al., 2012). Additionally, gossip can promote social change (Noon and Delbridge 1993) and its acceptance (Mills 2010), and can motivate self-improvement (Martinescu et al. 2014). Gossip also reinforces friendship (Ellwardt et al. 2012) and can enhance romantic relationships (Baxter et al. 2001).

Nevertheless, people treat gossip as malicious behavior (McDonald et al. 2007) and the gossip as evil or jealous (Einat and Chen 2012). Many studies support these conclusions. Gossips are perceived as more controlling and less emotionally warm than non-gossips (Farley et al. 2010). Active gossipers may not be trusted to keep discrete information to themselves, and therefore will be less attractive as close friends (Ellwardt et al. 2012). Robinson (2016) notes that gossip can arise as a desire for vengeance. Michelson and Mouly (2000) contend that it is largely motivated by the gossip's ego and competitiveness.

Moreover, most proverbs that relate to gossip warn against its negative consequences (Peters and Kashima 2013). Gossip thus receives pejorative labeling as a morally suspect activity. Gossip outcomes may create conflicts, suspicions, and encourages seeing the worst in other people (Michelson et al. 2010). Moreover, gossip reinforces stereotypes (Nikitina and Furuoka 2013). It registers a stronger impact on weak or disadvantaged people, and can cause great damage to a third party, even leading to social ostracism (Robinson 2016).

Leaper and Holliday (1995) found that for college students, negative gossip was more common than positive gossip. In our study, we attempted to evaluate people's reactions to a perceived indirect request to gossip: Would they do it willingly or reject the request? And if they assented, what would the content of their gossip look like - positive or negative?

2.2.Positive and negative gossip

Treating gossip as negative (or positive) raises the question of how to define this binary in the first place. Different classifications of gossip distinguishing between positive and negative types were thus surveyed. One classification sorted gossip into degrees of genuineness or falseness (Holland 1996). False gossip was perceived as negative and causing damage to others, while true gossip was perceived much more positively; for example, as a tool leading to better decisionmaking (Grosser et al. 2012). McDonald et al. (2007) found that if gossips are more intimate with each other, the gossip tends to be more genuine and honest. Another classification parsed a gossip's intentions: Do they mean to harm or benefit? Turner et al. (2003) support this claim that 'good gossip' is connected to trust and feelings of affection. Leaper and Holliday (1995) divided gossip into 'good', i.e. containing compliments or praise of a third party, and 'bad', i.e. consisting of criticism or derogatory remarks. However, a gossip's intentions do not always achieve the desired result. For example, Jackson (2012) determined type of gossip by its results. So-called 'good gossip' was a facilitator of a social process, while 'bad gossip' was harmful to the individual and group. Therefore, it is possible that an individual with unscrupulous intentions will facilitate the social process. Moreover, Robinson (2016) argues that outsider judgment of a gossip's intentions may be distorted. Kurland and Pelled (2000) divided gossip according to the positivity or negativity of the information. Thus, positive gossip is when the gossip deals with something good that happened to a third party, while negative gossip consists of bad news about someone other than the gossip. Watson (2011) suggested that people use positive gossip for approved behavior and negative for behavior lacking approval. The purpose of the gossip and the environmental perception of gossip also affect our judgment of gossip as good or bad (Luna and Chou 2013). Chua and De la Cema (2014) claimed that the nature of gossip is affected by the gossip's personality. For example, an aggressive gossip will gossip in order to achieve power and influence, with a greater tendency to gossip manipulatively and maliciously.

Grosser et al. (2012) focused on organizations, concluding that gossip can be simultaneously positive and negative. It ultimately depends on the perspective - is it the employee's perspective or the organization's? Gossip may benefit the individual but harm the organization and vice versa. Hafen (2004) made a similar distinction in the organizational domain, differentiating between gossip that benefits the organization and gossip harmful to it because it is only for individual profit. The difficulty to decide what is 'good' and what is 'bad' gossip led us to examine the traits that respondents mentioned in their descriptions

2.3.What are the traits that people are gossiping about?

Some scholars (for example, Waddington and Fletcher 2005, Palts and HarroLoit 2015) have investigated the emotions that people express when they speak about other people. For instance, it was found that in conversation about others, basic emotions such as anger, disgust, fear, sadness, and joy tend to be expressed, but not necessarily in a distinct sense (Greasley et al. 2000). Therefore, it is useful to isolate the most common forms of trait expression in gossip, however not many studies were conducted in order to uncover the trait expressions in gossip.

Salient traits in gossip are of great importance. For example, in one experiment participants had to decide how much money they wanted to transfer to a distant partner in a game, with participants receiving random gossipy information on the partner's level of stinginess or generosity (Sommerfeld et al. 2007). In the end, the distant partners presented as generous received far more money than those said to be stingy. This experiment sharpened the effects of gossip based on the usage of different traits. Roivainen (2015) found that in the Google Books corpus, the most popular attribute for others was 'intelligent'. From this finding, he concluded that intelligence is a central aspect of personality that affects the social status of a person.

Interestingly, it was also claimed that the more a person ascribes particular importance to a specific trait, the greater its prominence in language usage (Leising et al. 2014). They investigated 168 research participants to evaluate word usage when describing other people. Their findings indicate that in doing so, people prefer to use more general concepts such as 'good' or 'bad' rather than more specific traits.

Robinson (2016) claimed that many traits expressed in gossip are negative such as cheater, liar, or stingy. Personality traits of entrepreneurs were assessed as negative, including arrogant, manipulative, dramatic, and eccentric (Zibarras et al. 2008), while lower levels of traits connoting caution were used. However, Augustine et al. (2011) found that positive terms were generally used more often than their negative counterparts. Leising et al. (2014) supported this conclusion. However, Cohen (2011) found that negativity bias decreases when respondents are asked to describe a specific person. It was also found that positive words are more associated with relationships than negative words (Jose et al. 2010).

Hence, this study attempts to investigate if the most common traits that people mention are mostly negative (cf. Robinson, 2016) or positive (cf. Leising et al., 2014).

3.Method

As noted, qualitative methods were employed (Denzin and Lincoln 2000). These are more effective for locating meanings or interpreting phenomena in terms used by people in daily life. The information is elicited from participants directly and personally involved with these phenomena (Braithwaite et al. 2014). Qualitative methods are very useful in relationship research, as often the meanings people invest in relationships are paradoxical and contradictory (Manning and Kunkel 2014). As such, these methods offer deeper and profounder understandings of subjective meanings when individuals describe relationships (Fritz 2014).

We used Google Docs to contact as many respondents as possible. The question was distributed by third year undergraduate students to respondents which included other students, friends, and family They were asked to write down (in at least 4 sentences) how they would describe a third person to a friend and to indicate positive or negative traits relating to this individual. The word 'gossip' was not used in the request. After data collection, we had 2230 completed and relevant texts.

Respondent age ranged from 19 to 75, with a mean age of 30. Of the respondents, 953 (42.7%) were male and 1277 (57.3%) female. In order to validate our findings, we used two methods for analyzing the data. First, texts were categorized and analyzed traditionally using manual encoding by the two researchers separately. Then we compared the encodings and consolidated them until we arrived at established themes (Ryan and Bernad 2000). In qualitative research it is difficult to establish reliability, since striving for a common denominator may result in missing intermediate shades and thus losing the sensitivity of the instrument. Therefore, we first worked separately and after each researcher came to their own set of conclusions, we conferred to examine all similarities and differences in order to achieve the best results possible (Türk 2016). The study was conducted in Hebrew; in order to report its results, a translation and retranslation was carried out by two professional independent translators.

The traits were analyzed using a free-response design (Leising et al. 2014). In order to analyze such a large sample of text, software (TEXTIMUS) was specifically developed. This software assisted with text examination and analysis. We employed three of its major utilities.

The first utility (COUNT) generates sums for the number of appearances of every word using n-gram. In the field of computational linguistics, an n-gram is a contiguous sequence of n items from a given sequence of text (Davidovitch & Eckhaus, 2018). The items can consist of letters, words, or phonemes (Nadkami et al. 2011). An n-gram of size 1 is called a 'unigram', size 2 a 'bigram', and size 3 a 'trigram' - with larger sizes simply called 'n-gram' (He et al. 2009). For instance, applying n-gram on the sentence "human beings have the power to boost and sustain their lasting happiness" (Eckhaus & Sheaffer 2018a), unigram represent collection of single words and bigram every pair of words (such as 'human beings,' 'beings have,' 'have the,' etc.). This technique can also be applied for assessment of free text answers (Noorbehbahani and Kardan 2011), and was previously utilized to identify traits (Eckhaus 2016, 2017; Eckhaus & Sheaffer 2018b).

The second utility (ASSESS) takes as input an Excel sheet with a column of texts and a header row of selected words, and sums the number of each the words appearing for each of the texts. The module is able to match the words based on a wildcard; i.e., all words in which the selected word exists with or without a prefix or a suffix. This is particular useful for stemming, which is often employed in textanalysis (e.g. Eckhaus & Davidovitch 2018). For example, by choosing to count the word 'friend*' the software will count its derivatives: friend, friendly, friendlyness, etc.

The third utility (COMPARE) takes as input two texts and the required n-gram size, and generates two Excel files with the unique words used in both texts and a third Excel file with the words used in both texts.

In the first stage, we employed COUNT in order to detect recurring words or phrases. This step enabled identification of the most frequently recurring characterristic qualities which appear in respondent texts. Next, we defined a selection of the top frequent characteristics. We then checked if chosen words were also part of other words, and, if so, we selected these other words as well - to be later subtracted from the total count. For instance, the word 'good' is part of 'goods,' although 'goods' is not a trait. In the final count, therefore, 'goods' would be subtracted.

In the next stage, we employed ASSESS to the selection of chosen words in order to measure their frequency in each of the texts, and then we subtracted words with letters of selected words, but which were semantically unrelated. The module also allowed the examination of specific texts with high frequencies.

Finally, we employed COMPARE over several texts with high frequency of selected words in order to examine similarities and assess if further insight can be derived.

4.Results

4.1.Reactions to our request

4.1.1.Manner of compliance with request

As stated above, respondents were asked to describe to a friend the traits of a third person and to write down this description. The word 'gossip' was not mentioned in our short request. First, we looked at respondent reactions in their texts. Respondents drew on a wide range of interpretations of this request. They also employed several distinct strategies in registering their compliance and noncompliance with the request. Four response types were noted for the former:

1. Describing a person they know: Many of the respondents described a specific person that they met (although the request was very general and could equally apply to an imaginary person). For example, Respondent 339, who is a 28 year old female, wrote:

"L. is the worse manager ever; she just sits and commands people. She is giving orders and doing nothing except licking the management asses and that is the reason for her promotion."

It is clear that not only does this respondent know L - she has strong feelings towards her, which she desired to express. One explanation for this desire for specification in description may stem from the fact that it is easier to provide information on a known acquaintance. Additionally, a good opportunity to vent can be hard to pass up.

2. Outlining rather than engaging in description of a person: Some respondents resorted to the strategy of outlining how they would describe another person. That is, they did not, in fact, provide any specific details about the person. Instead, they attempted to avoid the request by detailing the descriptive process. For instance, Respondent 8 (a 38 year old male) wrote:

"At first I would say what I do not like briefly and without going into depth. Then I will present the traits that I like and at the end I will conclude in a tone of: 'But despite his negative traits, I think he is a good friend' or, alternatively, 'although this person can benefit me, I do not feel I can trust him.'"

Clearly, this respondent was not interested in describing a specific person. The answer was more of a subterfuge, possibly a means of avoiding gossip. Or alternatively, this respondent used this approach as he was unable to think of a specific individual. It is important to note that studies show that women and men engage in the same amount of gossiping activity (Eckhaus & Ben-Hador 2017).

3. Self-describing (or someone very similar): A few respondents declared either option openly, while others seemed to imply it with different levels of indirectness. For example, Respondent 85, a 41 year old male, conceded:

"This person is a lot like me; he loves to eat and spend time with his family and friends. Sometimes, he is nervous, but he does not hold a grudge and restrains himself easily."

This tendency is natural insomuch as people know themselves well and can easily resort to self-description. However, it may also serve as a gossip-avoiding technique. People may, in fact, feel more comfortable offering information about themselves than being perceived as gossips.

4. Describing an ideal person, usually from the opposite sex: For instance, Respondent 703, a 30 years old male, wrote:

"It is important that she will be a woman that will pamper and love you. She should be wise in the right degree and love children and childcare and, she should be warm and love to stay at home."

This description type may be connected to the life situation of the respondent. Wishful thinking of a desired partner may blend into the description request. In fact, description of an ideal person usually consists of only positive traits.

4.1.2.Refusal to gossip

Approximately, a third of the respondents referred to our request directly as a request to gossip. As noted in the literature review, people tend to refer to gossip as malicious and harmful. Indeed, most respondents who wrote the word 'gossip' registered opposition to it (about 75% from them-=558 respondents). These gossip opponents expressed their lack of compliance in several ways - from tacit to direct refusal. Refusal types were categorized into four different responses:

1. Explicit refusal: Some respondents openly declared their refusal to gossip. For example, Respondent 1311, a 23 year old female, stated:

"Nobody is perfect and we shouldn't judge people without a reason; people must stop gossiping at others and start looking at themselves."

As mentioned, no direct request to gossip (or pass judgment) was made. This reaction, however, emphasizes that our request was indeed perceived as an invitation to gossip. Most opponents in this category attempt to justify their disapproval of gossip. One example is Respondent 514, a 33 year old female, who expanded on the social implications of gossip:

"I do not think I would define any one; never should we judge another person. Once we understand this, we as a society will look totally different and will not need to gossip or judge a person, and should deal only with significant issues."

This respondent frames her objection to a perceived gossip request by articulating her general point of view about gossiping. She clearly views gossiping as detrimental to the fabric of society and therefore any perceived request to gossip immoral.

2. Angry refusal: Some respondents expressed anger towards the researchers for asking them to engage in such immoral/unworthy behavior. A typical reaction is exemplified by Respondent 1062, a 60 year old male:

"I think talking about other people is rude!"

This reaction is similar to the previous one, although with an added note of reproof of the researcher. The respondent here pointedly accuses the researchers of rudeness.

3. Reference to gossip as a despised trait: Some respondents focused their ire on the perceived negativity of gossiping. For instance, Respondent 1823, a 27 year old male, referred to gossip as:

"An attribute that I don't like - gossip and transferring information from one person to another; you cannot trust a gossiper because he doesn't tell the truth."

Instead of complying with the request to describe another person, this respondent expressed his view of gossip as a negative trait. That is, he was employing a tactic to oppose the request and reprove the researchers (as with the previous example).

4. Sublimation of refusal: The last type of opposition was the most subtle. These respondents sublimated their answer and used only positive traits. For instance, Respondent 1856, a 52 year old female, wrote:

"An interesting person, sympathetic, tactful, honest and does not apologize for who he is."

Although the respondent was asked to use both positive and negative traits, she selected exclusively good qualities. Moreover, some respondents even declared that they would only employ positive traits.

4.2.Prominent features mentioned when gossiping

In order to rate the positive and negative traits that appeared in the text, we used the TEXTIMUS software. 4222 traits were coded. Many traits appeared in different terms such as fibster and liar, Therefore some of the categories were grouped, the grouping was made by two independent judges. The most common positive traits that were found are:

1. Good - including good person, good human being, good man, good woman, good heart: 803 appearances, in total.

2. Friend - Good friend, friendly: 577 appearances.

3. Nice - including pleasant, kind, cute: 345 appearances.

4. Smart - including witty: 343 appearances

5. Help - Care, aid, support: 221 appearances.

6. Happy -Fun, funny, sense of humor: 199 appearances.

The most common negative traits (after grouping) are:

1. Arrogant: including pride, selfish, arrogance: 178 appearances.

2. Stingy: 58 appearances.

3. Liar: 57 appearances.

4. Stubborn: 51 appearances.

5. Lazy: 29 appearances.

Figure 1 presents a word cloud. Word size reflects frequency of each word.

4.3.Analysis of results

It can be concluded that even though the word 'gossip' was omitted from the request, many respondents perceived it as an invitation to gossip.

Respondent reactions can be divided into two categories: First, we found respondents who cooperated with the researchers and engaged in the description of the other person. Respondents interpreted this request variously - describing a person they know, someone they would like to meet, self-description, and metadiscourse of how they would describe another person.

The second category consisted of respondents more aptly described as opponents. They resisted the request to describe the other person and this failure to comply was interpreted as a rejection of gossip. However, this resistance took different shapes - from tacit resistance reflected by willingness to describe only positive traits, to active resistance expressed in preaching or rebuking.

Moreover, cooperation of participants from the first category should not be assumed to be unconditional. For example, description of an ideal person is very similar to willingness to describe using only positive traits. The difference can be marginal and hard to detect. Those who describe themselves mostly favored positive traits, although some included self-criticism. As noted, self-description can be an elegant subterfuge to avoid describing someone else (i.e. gossiping). Meta-description (i.e. outlining the process of describing someone without using traits related to a specific individual) is also a convenient instrument for nonconfrontational avoidance of the request. As such, most respondents who privileged this strategy refrained from using negative traits.

The trait count leads to the same conclusion - respondents generally prefer to use positive traits. Positive traits total is 3168 occurrences in the texts, whereas the total for negative traits is 1054. The latter thus amount to approximately a quarter of the general traits word list. As noted in the literature review, much research points to the benefits of gossip. Nevertheless, most respondents actively or passively opposed the request to write down what they perceived as gossip.

It is noteworthy that the most prominent trait that respondents use to describe someone else is 'good'. This can be employed adjectively (i.e. a good athlete), but is also arguably the ultimate positive trait and is used generously. Interestingly, regardless of the very general request to describe a person using traits, respondents hardly used physical and appearance-based traits. Only 1.5% of all traits analyzed related to appearance, and most were positive (e.g. beautiful).

5. Discussion

The results showed that respondents are opposed to what they perceive as gossip, preferring more positive descriptions. Watson (2011) suggested that the individual's relationship to gossip can be a reflection of inner life and unconscious processes. Massa and Simeoni (2014) posit that storytelling and gossip are the primary modalities through which people construct, interpret, and experience reality. Gossip seems to be a frequent and active behavior, with Emler claiming that 66% of conversational duration devoted to gossip (forthcoming). In real life, however, negative gossip is far more prevalent than in studies (Turner et al. 2003). Foster (2004) refers to this incongruity as 'the paradox of gossip'.

The gap between intention and behavior (Fishbein and Ajzen 1975), especially in gossip, has been investigated and documented. Luna and Chou (2013) found that attitude and subjective norms are significant predictors of intention to gossip, but not to actual behavior. A compelling explanation for this gap can be attributed to two notions - positivity bias (i.e. the Pollyanna Hypothesis) and perception of gossip as evil.

The Pollyanna hypothesis, a form of positivity bias, claims that speakers prefer to use more positive than negative words. Many studies have confirmed its validity, especially those in which participants were asked to describe others (Augustine et al. 2011). For example, Leising et al. (2014) isolated terms used to describe individuals and found that people use significantly more positive than negative terms. It was also shown that positive emotional bias features quite prominently in written expressions (Garcia et al. 2012). In stories and writings, people are predisposed to prosocial communication and, therefore, use positive terms (Kloumann et al. 2012).

Boucher and Osgood (1969) coined 'the Pollyanna hypothesis', a reference to the heroine of the book, Pollyanna. Pollyanna is characterized by her unquenchable optimism and positive approach to every trouble. Carr (2011) defined the Pollyanna effect as the bias of people to view themselves in the best possible light. The human mind is 'programmed' to think positively rather than realistically. As such, people resist perceiving themselves as gossipers.

Hildebrandt and Snyder (1981) extend the Pollyanna effect beyond selfperception to the optimistic framing of events and other people. It is also contended that this bias is particularly well-expressed in writing (Kloumann et al. 2012). Hence, we can conclude from the Pollyanna hypothesis that most people will be biased to write positive traits based on deep cognitive structures.

The Pollyanna bias may have impacted respondent descriptions. But we also postulate that perception of our request as an invitation to 'gossip' was also affected by entrenched cultural notions of gossip as evil speech.

Social norms (Ben-Hador, 2017) mandate that gossip is perceived as evil. As noted, some of our respondents reprimanded the researchers due to their perception of our request as an invitation to gossip. Gossip is condemned as a negative trait in many religions and cultures. Capps (2012) pointed out how Christianity prohibits gossip as evil. Jewish religious law warns its believers against gossip (Buddenbaum 2014). Similar prohibitions are found in Islam (e.g. Qur'an, Surah 49: verse 12). Naturally, people avoid perceiving themselves as engaging in activities forbidden by tradition and transgressing religious precepts. Community acceptance is vital for the individual in social groups and gossiping is socially stigmatized (Einat and Chen 2012). According to Hartung and Renner (2013), people tend to not perceive themselves engaging in gossip, but are quick to believe others are gossipers. When people collect information about others they rationalize it as mere selective paying of attention. People deny involvement in gossip just as they deny other bad habits (Carr 2011). In sum, the perception of gossip as evil is a social perception - norms rooted in religion and ancient customs.

By using mainly positive descriptions, speakers can convince themselves that they do not really gossip. Luna and Chou (2013) claimed that gossips are perceived as on a higher moral level if the content of their gossip is positive. In our study, the specific request to record descriptions in writing was designed to afford respondents time for careful deliberation. We postulated that they would then write down only words that flattered their self-perception. In contrast, free form speech would have consisted of many 'slips of the tongue'. Writing, however, offers the possibility of self-editing and revision as the writer seeks optimal forms of expression.

Regarding the prominence of the trait 'good', Skinner (1972) claimed that the trait 'good', represents a very broad concept. Garcia et al. (2012) supported this claim, adding that positive words like 'good' denote a higher level of generality, but are also coded with less information. Still, Skinner (1972) suggested that the word 'good' is used to describe things which almost everyone values. The behaviors surrounding this value are reinforced and thereby contribute to cultural survival. That is, people may gravitate towards usage of 'good' in their descriptions in order to feel like better human beings (Garcia et al. 2012). Additionally, respondents avoid relating to the externality of other people and to their appearances so as to generate the same sentiment of feeling like a better human being. As such, the first (i.e. people prefer to avoid gossiping) and second part of the analysis (i.e., significance and number of trait selection in description) support each other.

Nevertheless, counting traits may limit this study as our research uses qualitative methods because it implies losing depth at the base of qualitative research and the ability to reach compound conclusions. We tried to overcome this limitation by using classic qualitative analysis alongside the traits counting. Moreover, traits were analyzed using a free-response design - open choice of words and terms is uncommon in traits studies, and usually the traits are presented in a list supplied by researchers. As such, this study design creates an advantage that stems from reinforcement of the ecological validity of our research as compared to advance selection of terms (Leising et al. 2014).

Another limitation of this study is that intentions were evaluated through selfreporting and not analysis of actual behavior. Miller's (1961) negativity bias suggests that effect of positive descriptions on behavior will be limited. Ultimately, what matters is actual individual speech behavior, and, therefore, more research is needed to inspect real gossiping behavior.

Sidebar

Addresses:

Eyal Eckhaus

Department of Economics and Business Administration

Ariel University

Ariel 40700, Israel

E-mail: [email protected]

Batia Ben-Hador

Department of Economics and Business Administration

Ariel University

Ariel 40700, Israel

Email: [email protected]

References

References

Augustine, A. A., M. R. Mehl, and R. J. Larsen (2011) "A positivity bias in written and spoken English and its moderation by personality and gender". Social Psychological and Personality Science 2, 508-515.

Baxter, L. A., T. Dun, and E, Sahlstein (2001) "Rules for relating communicated among social network members". Journal of Social and Personal Relationships 18, 173-199.

Beersma, B. and G. A. Van Kleef (2012) "Why people gossip: an empirical analysis of social motives, antecedents, and consequences". Journal of Applied Social Psychology 42, 26402670.

Ben-Hador, B. (2016a) "Coaching executives as tacit performance evaluation: a multiple case study". Journal of Management Development 35, 75-88.

Ben Hador, B. (2016b) "How intra - organizational social capital influences employee performance?" Journal of Management Development 35, 1119-1133.

Ben Hador, B. (2017) "Three levels of organizational social capital and their connection to performance". Journal of Management Development 36, 348-360.

Boucher, J. and C.E Osgood (1969) "The pollyanna hypothesis". Journal of Verbal Learning and Verbal Behavior 8, 1-8.

Braithwaite D. O. Abetz. J. and J. Moore (2014) "'I need numbers before I will buy it': reading and writing qualitative scholarship on close relationships". Journal of Social and Personal Relationships 4, 490-496.

Buddenbaum, J. M. (2014) "God, Jews and the media: religion and Israel's media". Shofar 32, 137139

Capps, D. (2012) "Gossip, humor, and the art of becoming an intimate of Jesus". Journal of Religion and Health 51, 99-117.

Carr, A. (2011) Positive psychology:the science of happiness and human strengths. London; New York: Routledge.

Chua, S. V. and K. J. De la Cerna Uy (2014) "The psychological anatomy of gossip". American Journal of Management 14, 64-69.

Cohen, S. J. (2011) "Measurement of negativity bias in personal narratives using corpus-based emotion dictionaries". Journal of Psycholinguistic Research 40, 119-135.

Cole, J. M. and H. Scrivener (2013) "Short term effects of gossip behavior on self-esteem". Current Psychology 32, 252-26.

Cuonzo, M. A. (2008) "Gossip: an intention-based account". Journal of Social Philosophy 39, 131140.

Davidovitch, N. and Eckhaus, E. (2018) "The influence of birth country on selection of conference destination-employing natural language processing". Higher Education Studies, 8(2), 92-96.

Denzin, N. K. and Y. S. Lincoln (2000) "Introduction: the discipline and practice of qualitative research". In N. K. Denzin and Y. S Lincoln, eds. Handbook of qualitative research. 2nd ed., 1-44. Thousand Oaks: Sage Publications.

DiFonzo, N. and P. Bordia (2007) Rumor psychology: social and organizational approaches. Washington, DC: American Psychological Association.

Dunbar, R. (2008) Grooming, gossip, and the evolution of language. Cambridge: Harvard University Press.

Eckhaus, E. (2016) "Corporate transformational leadership's effect on financial performance". Journal of Leadership, Accountability and Ethics 13, 90-102.

Eckhaus, E. (2017) "A shift in leadership". Academy ofStrategic Management Journal 16, 19-31.

Eckhaus, E. and Ben-Hador, B. (2017) "Gossip and gender differences: a content analysis approach". Journal of Gender Studies, Publised online at: https://doi.org/10.1080/09589236.2017. 1411789, 1-12.

Eckhaus, E., and Sheaffer, Z. (2018a) "Happiness enrichment and sustainable happiness". Applied Research in Quality of Life. doi:https://doi.org/10.1007/s11482-018-9641-0

Eckhaus, E., and Sheaffer, Z. (2018b) "Managerial hubris detection: the case of Enron". Risk Management. Published online at https://doi.org/10.1057/s41283-018-0037-0.

Eckhaus, E. and Davidovitch, N. (2018) "Improving academic conferences - criticism and suggestions utilizing natural language processing". European Journal of Educational Research,7, 445-450.

Einat, T. and G. Chen (2012) "Gossip in a maximum security female prison: an exploratory study". Women and Criminal Justice 22, 108.

Ellwardt, L., C. Steglich, and R. Wittek (2012) "The co-evolution of gossip and friendship in workplace social networks". Social Networks 34, 623-633.

Emler, N. (forthcoming) Serpent's tongue: the social psychology of gossip. Oxford: Oxford University Press.

Farley, S. D., D. R. Timme, and J. W. Hart (2010) "On coffee talk and break-room chatter: perceptions of women who gossip in the workplace". The Journal of social psychology 150, 361-36.

Fishbein, M. and I. Ajzen (1975). Belief, attitude, intention, and behavior: an introduction to theory and research. Reading, MA: Addison-Wesley.

Foster, E. K. (2004) "Research on gossip: taxonomy, methods, and future directions". Review of General Psychology 8, 2, 78-99.

Fritz, J. H. (2014) "Researching workplace relationships: what can we leam from qualitative organizational studies"? Journal of Social and Personal Relationships 31, 460-466.

Garcia, D., A. Garas, and F. Schweitzer (2012) "Positive words carry less information than negative words". Data Science 1, 3, 1-12.

Greasley, P., C. Sherrard, and M. Waterman (2000) "Emotion in language and speech: methodological issues in naturalistic approach". Language and Speech 43, 355-357.

Grosser, T. J., V. Lopez-Kidwell, G. Labianca, and L. Ellwardt (2012) "Hearing it through the grapevine: positive and negative workplace gossip". Organizational Dynamics 41, 52-61.

Hafen, S. (2004) "Organizational gossip: a revolving door of regulation and resistance". The Southern Communication Journal 69, 223-240.

Harari, Y. N. (2014) Sapiens: a brief history of humankind. New York: Harper.

Hartung, F. and B. Renner (2013) "Social curiosity and gossip: related but different drives of social functioning". PLoS One 8, 1-10, e69996.

He, Y., S. C. Hui, and T. T. Quan (2009) "Automatic summary assessment for intelligent tutoring systems". Computers and Education 53, 890-899.

Hildebrandt, H. W., and R. D. Snyder (1981) "The Pollyanna Hypothesis in business writing: initial results, suggestions for research". Journal of Business Communication 18, 5-15.

Holland, M. G. (1996) "What's wrong with telling the truth? An analysis of gossip". American Philosophical Quarterly 33, 197-209.

Jackson, C. (2012) "Using gossip constructively as a part of holistic process". Holistic Nursing Practice 26, 183-187.

Jose, A., S. Rajaram, K. D. O'Leary, and M. C. Williams (2010) "Memory for partner-related stimuli: free recall and frequency estimation". Journal of Social and Personal Relationships 27, 658-670.

Kloumann, I. M., C. M. Danforth, K. D. Harris, C. A. Bliss, and P. S. Dodds (2012) "Positivity of the English language". PLoS One 7, e29484.

Kurland, N. B. and L. H. Pelled (2000) "Passing the word: toward a model of gossip and power in the workplace". Academy of Management Review 25, 428-438.

Leaper, C. and H. Holliday (1995) "Gossip in same-gender and cross-gender friends' conversations". Personal Relationships 2, 237-246.

Leising, D., J. Scharloth, O. Lohse, and D. Wood (2014) "What types of terms do people use when describing an individual's personality"? Psychological science 25, 1787-1794.

Luna, A. and S. Y. Chou (2013) "Drivers for workplace gossip: an application of the theory of planned behavior". Journal of Organizational Culture, Communications and Conflict 17, 115-129.

Manning, J. and A. Kunkel (2014) "Making meaning of meaning-making research: Using qualitative research for studies of social and personal relationships". Journal of Social and Personal Relationships 31, 433-441.

Martinescu, E., O. Janssen, and B. A. Nijstad (2014) "Tell me the gossip". Personality and Social Psychology Bulletin 40, 1668-1680.

Massa, E. and V. Simeoni (2014) "Once upon a tale. On the foundational role of narrative in constructing linguistic and social experience". Journal of Comparative Research in Anthropology and Sociology 1, 75-100.

McDonald, K. L., M. Putallaz, C. L. Grimes, J. B. Kupersmidt, and J. D. Coie (2007) "Girl talk: gossip, friendship, and sociometric status". Merrill-Palmer Quarterly 53, 381-411.

Michelson, G. and V. S. Mouly (2000) "Rumour and gossip in organisations: a conceptual study". Management Decision 38, 339-346.

Michelson, G., A. van Iterson, and K. Waddington (2010) "Gossip in organizations: contexts, consequences, and controversies". Group and Organization Management 35, 371-390.

Miller, N. E. (1961) "Some recent studies of conflict behavior and drugs". American Psychologist 16, 12-24.

Mills, C. (2010) "Experiencing gossip: the foundations for a theory of embedded organizational gossip". Group and Organization Management 35, 213-240.

Nadkarni, P. M., L. Ohno-Machado, and W. W. Chapman (2011) "Natural language processing: an introduction". Journal of the American Medical Informatics Association 18, 544-551.

Nikitina, L. and F. Furuoka (2013) "Dragon, kung fu and Jackie Chan": stereotypes about China held by Malaysian students". Trames 17, 2, 175-195.

Noon, M. and R. Delbridge (1993) "News from behind my hand: gossip in organizations". Organization Studies 14, 23-36.

Noorbehbahani, F. and A. A. Kardan (2011) "The automatic assessment of free text answers using a modified BLEU algorithm". Computers and Education 56, 337-345.

Palts, K. and H. Harro-Loit (2015) "Parent-teacher communication patterns concerning activity and positive-negative attitudes". Trames 19, 2, 139-154.

Peters, K. and Y. Kashima (2013). "Gossiping as moral social action: a functionalist account of gossiper perceptions". In J. Laszlo, J. Forgas, and O. Vincze, eds. Social cognition and communication. New York: Psychology Press.

Peters, K. and Y. Kashima (2015) "Bad habit or social good? How perceptions of gossiper morality are related to gossip content". European Journal of Social Psychology 45, 784-798.

Robinson, B. (2016) "Character, caricature, and gossip". The Monist 99, 198-211.

Roivainen, E. (2015) "Personality adjectives in twitter tweets and in the google books corpus- an analysis of the facet structure of the openness factor of personality". Current Psychology 34, 621-625

Rosnow, R. L. (1988) "Rumor as communication: a contextualist approach". Journal of Communication 38, 12-28.

Ryan, G. W. and H. R. Bernard (2000) "Data management and analysis methods". In N. Densin and Y. Lincoln, eds. Handbook of Qualitative Research. 2nd ed., 769-802. Thousand Oaks, CA: Sage Publications.

Skinner, B. F. (1972) Beyond freedom and dignity. New York: Bantam Books.

Sommerfeld, R. D., H.J. Krambeck, D. Semmann, and M. Milinski (2007) "Gossip as an alternative for direct observation in games of indirect reciprocity". PNAS 104, 17435-17440.

Türk, K. (2016) "Performance management of academic staff and its effectiveness to teaching and research - based on the example of Estonian universities". Trames 20, 17-36.

Turner, M. M., M. A. Mazur, N. Wendel, and R. Winslow (2003) "Relational ruin or social glue? The joint effect of relationship type and gossip valence on liking, trust, and expertise". Communication Monographs 70, 129-141.

Waddington, K. and C. Fletcher (2005) "Gossip and emotion in nursing and health-care organizations". Journal of Health Organization and Management 19, 378-394.

Watson, D. C. (2011) "Gossip and the self". Journal of Applied Social Psychology 41, 1818-1833.

Zibarras, L. D., R. L. Port, and S. A. Woods (2008). "Innovation and the 'dark side' of personality: dysfunctional traits and their relation to self-reported innovative characteristics". Journal of Creative Behavior 42, 201-21.