Gender Difference in Hedging: A Corpus-Based Study to TED Talks about Emotion

In the development of linguistics, many experts tried to identify the gender difference in language use from different perspectives. Many studies show that different genders have different preference when choosing hedging words to express their stands. But recently some other researches doubted the existence of such gender difference in pragmatic. This study established a corpus and analyzed 12 TED talks delivered by different genders and tried to find out whether the gender difference in hedging words exist. The result shows that gender difference does exist in word choice and lexical diversity. This study may shed some light for the following study to continue the research of gender difference from the angle of word choice.

focuses on the relationship between language and social reality.
For tertiary level students, learning how to achieve learning goals in a cooperative mix-gender environment is beneficial. Be aware of the gender difference of using words and communicative strategies can facilitate the students in setting up a humongous learning context and enhance their own performance in academic tasks. Hedging is one of the most commonly used meta discourse strategies in academic genre (Aull & Lancaster, 2014), which includes a number of linguistic forms used to reduce the aggressiveness and increase the discursive space. TED talks are popular academic spoken genre which allowed the presenters to express their innovative ideas to the public in a talk. Most speakers in TED are experts or professionals in certain fields, their talks represented most features in academic speaking. Researchers reported TED talks as a beneficial resource for English language learning not only in listening but also in speaking (Banker & Gournelos, 2013). Learning the use and difference of hedging words in TED talks can facilitate students' understanding of the rhetorical nature of academic spoken genre.

Literature Review
Gender difference in language Gender difference in language use is somehow pre-assumed perception for many years, Canary & Hause (1993) once argued the reason for researching gender difference in communication. After summarizing the result from 10 metaanalysis in concerned about the gender difference in various linguistic topics the authors drew a conclusion that the difference between gender is seemed to be "muddled and negligible", for the reason of using invalid measure instruments, inconsistent ideas of distinction between sex and gender as well as the lack of supportive theories to address the cause of the difference. Although as the end of the paper Canary and Hause claim that there was no necessity to study the gender difference in communication for the small effect size of gender, many other studies proved that in the choice of linguistic form including hedging words the gender-difference do exist. For the past two decades many researchers conducted studies to explore the difference between men and women in using linguistic forms cross genre of written and spoken (Samar & Alibakhshi, 2007;Holmes, 1986).

Rhetorical strategy of hedging
Hedging words represent a variety of linguistic devises used to reduce epistemic commitment and increase discursive space. Studies claimed that the use of hedging words is recognized as appearance-based phenomenon with various categories (Aull & Lancaster, 2014). Hedges are regarded as an effective strategy in communication for negotiating ideas across context. The purpose of using hedging is helping the readers or audience to understand the meaning, in both spoken and written language. In the academic genre writers tend to use hedges to provide a suitable representation of their ideas (Hassani & Dastjani Farahani, X. W. Li, F. P. Li 2014). As well as in casual interactions, hedging words are employed as a method for turn taking, conveying politeness, reducing face-threats and indicating proper vagueness. Many researchers found that hedging words are discipline specific and influenced by many other factors like purpose and styles (Hyland, 1995). Hyland (1995) defined hedging as the "expression of tentativeness and possibility in language use" it is a salient feature in the academic genre for the "statement is rarely made without subjective assessment of truth." (P33). In spoken discourse hedging is used to indicate to what extent the speaker is confident about his/her ideas. In an academic talk the most important part is to convey presenter's ideas by interpreting the data or observations in the study. The process is affected by the essential elements of communication procedure, speaker, audience, language, and context. The understanding of knowledge requires the joint efforts from both the speaker and the audience through rhetorical methods of persuasion. The hedging words here tend to serve the role of accomplishing the successful goal of understanding.
Gender difference in hedging Studies about the relationship between gender and hedging is somehow influenced by Robin Lakoff (1975). In her book she discussed the phenomenon that women's speech seemed to be less aggressive for the reason of being more "feminine" and tend to employ less assertive strategies in communication. Lakoff used the phrase "women language" to address a series of devises serving the function. Hedging is one of the ten features of women's language which is used to indicate the probability and certainty of agreement, it is a common strategy in academic writing and discourse. Newman et al. (2008) investigated 14,000 texts through a corpus-based method and found the tendency that women tend to use more hedging words in their writing to indicate politeness which was consistent with previous findings.
Hassani & Dastjani Farahani (2014) also explored the gender difference of using hedging in research articles in the discipline of applied linguistic in Iran. They found that use of hedges was influenced by discipline, culture and language competence, what's more gender also plays an important role in the choice of hedging devices. They studied the hedging words which were categorized based on Hyland's (1996) model in 60 research articles and finally concluded that the gender difference in hedging words is significant for men use more hedging words than women did which was totally contrast with the findings from Lakoff (1975).
In the spoken genre, Albaqami (2017) did a multi-modal text analysis on the different use of hedging devices between gender, the findings showed that the difference between men and women is obvious. The purpose for using hedges altered between genders, women use these words to avoid expressing their ideas directly for the reason of feeling insecure and lack of power in the conversation.
Contrarily, men used hedges primarily to address specific purpose like enhanc-X. W. Li, F. P. Li ing their position in order to control the debate. Mostofee & Pour (2016) focused on the gender difference of using hedges in classroom learning. the authors designed different types of oral discussion tasks in a Iranian English classroom and recorded the frequency of using hedging words when talk with same and mixed gender partners. The results indicated that women tend to use more hedging words as a communicative strategy when discussing with same gender partners similar with the result when they talk in a mixed gender groups.
Findings from studies conducted in different context about different genre did reveal consistent result about the gender difference in using hedging words. In order to make the gender distinction more explicit, more detailed study should be conducted to provide more evidence to address to difference. So, in this study, the author will try to explore the difference of using hedges between women and men in the genre of TED talks about emotion, the research questions are: 1) Is there difference between gender of using hedging strategy in TED talks about emotion in terms of frequency?
2) Is there gender difference in the preference of using particular hedging expressions in different categories of hedging?

Data collection
As Hyland (1995) mentioned that hedging words allow the writer to express the possibility or attitude to the truth, and cross genre research found that hedging words more appeared in the socio-science genre for the interpretation of the result based more on the presenters' subjective personal observation and experience. In order to including more hedging words for analysis we chose the emotion section in TED talks as the research genre. For interpretation of emotion is subjective, the presenter need to use hedging to reduce the aggressiveness of his/her findings and make it acceptable for the audience. This section contains 13 talks from 6 female and 7 male presenters, for the reason of equality in number the author deleted the least viewed talk in the section which is presented by a man. The reason to list the talks in the order of viewing rate is that the rate indicated the preference of the audience and popularity of the video which is a good evidence for the quality of the talks. Talks with high quality should be good in both the content and the structure, so the words used in high quality talks can be more representative of the features in this genre. For the purpose of applying the materials in the English classroom, all the selected talks are delivered in English.
And the presentation should be delivered mainly by words, talks include other form of presentation like music and dance were excluded. The transcripts of the 12 talks are provided by the TED website, the author listened to all the talks and checked the accuracy of the transcripts as well.
Classification of hedging words Many researches revealed that hedging is mainly realized through appearance Most of the categories are easy to understand except approximative hedges refers to those hedging words which are used to indicate the degree to which a claim is true or acceptable. The current study will investigate gender difference of using these four types of hedging words in TED emotion talks. The author used wordsmith to identify the descriptive data of the two groups of talks try to identify some gender difference in length of talk, high frequency words and the variety of vocabulary.
TED talks are classified into different categories according to the topics. The author chose 12 talks with the tag of emotion for the reason that emotion is the most discussable and complex topic between the two genders. Speakers who deliver talk about this topic may use more hedging words to show their stands. The author selected the top six talks with the most view times in both genders. After checking, all the 12 texts were uploaded to the wordsmith for further research, the author used concord to search for the concordance contained words or phrases in the hedging word list, for the size of current corpus is small (approximately 25,000 words), the author didn't set the minimum frequency when counting. The raw frequency was recorded and normalized for further comparison. Wordsmith can only provide frequency for single word so the frequency of those two and three words hedges was counted by the author through checking the concordance lines. The result is discussed in the following section.

Findings
Word choice and lexical diversity As a powerful word processing software wordsmith can provide wordlist of a corpus as well as basic statistic information. From the following chart we can see that the mean type/token ratio of man and woman TED talks is 35.67 (man) versus 39.86 (woman) as showing in Table 1. The author ran independent T-test by using SPSS proved that the difference of STTR between genders is marginally significant (p = 0.05, d = 1.23). The result indicates that women use a larger variety of words in expressing their opinions when comparing with men, and the large effect size indicate the difference is noticeable. This result seemed prove that the difference in diversity of words choice between gender do exist. The possible explanation for not significant difference may due to the small size of current which only contain 12 talks with total number of 25,000 words, if the corpus contains more talks like 30, the result should be significant.
Result of hedging words In this study the hedging words are classified into four categories the raw and normalized frequency provided by wordsmith are listed below, some tendency can be seen from the data.
From the result illustrated in Table 2, generally speaking there is a slight difference between men and women in terms of hedging, and a tendency can be found that the for the most frequently used two types of hedging namely approximative hedges and modal hedges men and women seem to have no difference if take the frequency into consideration. The more salient difference can be seen in there other two less frequently used hedging category, self-mentioned hedges and evidential verb hedges with men almost use twice as many as these two types than women. The possible explanation will be presented in the discussion section.
Preference hedging expressions in different categories After normalization, the author listed the top three hedging in the four categories in terms of frequency in both genders, the result in Table 3 indicated that men and women do have different preference when choosing hedging words or phrases in the four categories. What is interesting is that the most frequently used hedging in all the categories are the same cross gender, with the following  two are different. Except for the self-mention hedges. Because of the small size of the corpus in the category of self-mention hedges only two phrases can be found in the 12 talks so the comparison in the type cannot be made.

Discussion
Although Lakoff's (1975) book Language and woman's place is considered as the masterpiece in the study of hedging and gender and determining that hedging is one of the ten features of "women language" therefore this strategy is more commonly used by women in both written and spoken genres. Many other researchers have conducted relevant studies in different disciplines to explore the real meaning of gender difference in language use and concluded that the gender difference of using hedging in academic genre may vary across discipline, for content, purpose of communication, language proficiency, and culture can also influence the use of hedging in speaking. Findings from this study provided evidence for the gender difference of hedging use is strongly affected by the topic, content of speech and audience, so when men are talking about sensitive and subjective topics like emotion in front of common and general audience they tend to use more hedging strategy to reduce their aggressiveness and hardness.
Male speakers try to use hedging to establish more empathetic and sympathetic image to the audience, for men are usually considered as less emotional and less sensitive about emotion than women, being aware of the repertoire in the audience, using more hedging can soften their masculine unreliable concept therefore increase their credibility in addressing such topic. In contrast women speakers built up more confident by adopting a more male like strategy of using less hedging word in avoiding to activate audience convention of women language and hope they could by viewed as professional presenters on a TED con-ference stage. And the confidence of clearer and more subtle understanding of emotion may enhance female speakers more assertive expression in their presentation. The purpose of TED talk is to spread the innovative ideas to worldwide audience. In order to achieve persuasiveness and leave a deep impression to the audience women presenters may tend to break the pre-assumed image and influence the listeners with not only the provoking propositions but also a unique self-image.
As a kind of human activity, realization of hedging strategy is not influenced by single factor, other research has provided evidence (Hegde, 2017) to show that rhetorical strategies like hedging can be affected by multiple factors, like gender, audience, topic, content, and culture as well as personal stylistic. The interaction among all those factors made a complicated picture of determining the effective size of single factor, in such condition it is difficult the identify the effect of gender. Future study may take the interaction among factors into consideration and multimodality analysis should be conducted.
According to the definition provided by Nivales (2011) andHyland (1995), self-mention hedges refer to the hedging expressions include first person pronoun or mentions about the author. This kind of hedging indicates the proposition relation between the statement and presenter, in contrast evidential emphasis the impersonal relation and highlight the objectiveness of the claim, both these two categories seemed to be more powerful than the other two types. The result in this study illustrated that men used far more hedging in these two sections than women, which may to some extent be identical to the pervious findings that the purpose of using hedging differs between men and women (Albaqami, 2017). For men use hedging to enhance their power in presentation and women tend to use it for reason of avoiding impoliteness. So men speakers used more of the hedging in the powerful categories and women chose hedging from the other two more neutral categories.

Implication and Limitations
This study tried to explore whether the gender difference in hedging really exist by conducting a small corpus-based analysis to the emotion TED talks. From the socio-cultural linguistic perspective, the constant question asked by researchers is whether men and women behave differently when using language. Gender-based difference and the learning environment can influence students' language learning outcomes (Mostofee & Pour, 2016). Many studies proved that gender difference may affect the particular way of using the language interaction between male and female students and in turn affect their performance. The result provides some evidence to show the tendency that to some extent there is gender difference in the frequency of using hedging, but a clear conclusion cannot be draw without further study in relevant disciplines in TED talks. TED talks as one of the popular academic spoken genre can provide sufficient examples for the students understanding the gender difference of using hedges in academic spoken context, it is a very effective and appropriate resource which can be used by the teachers to facilitate students improvement in English proficiency and guide students to be skilled communicators by using proper rhetorical strategies when communicate with different people in different environment. To know the gender difference in using hedging words also can help the teachers pay more attention to students' performance in mixed class. Teachers should provide different materials or inputs for different students. Awareness of gender difference can facilitate the students catch more detailed information in the talks and improve their skill of how to appreciate English talks. The small size of the corpus which only contained 12 talks and approximately 25,000 words may affect the reliability and validity of the result, so the author cannot draw a concrete conclusion from this study, but the result do reveal a tendency that some king of difference do exist between gender, but further studies should be conducted with a larger sample size for a clearer determination of the gender difference in hedging.
The findings in this study was merely concerned about one of the disciplines in TED talks future studies should be conducted to investigate whether this pattern of gender difference in hedging is commonly occurred cross discipline or is it a discipline specific phenomenon which may vary according to different topics. Other factors like language proficiency and culture influence which may affect the different use of hedging in TED talks also should be taken into consideration. The finding showed the importance of guiding the students to clearly understand the factors which affect language use in oral presentation and emphasis the importance of acquiring the knowledge of gender difference in language use for becoming skilled speakers in English.