In this section, we will examine whether the severity of an event affects the participants’
willingness to give advice and choices of advice types. Section 4.3.1 presents the overall findings and Section 4.3.2 discusses the interaction of severity with other factors.
4.3.1 Overall Findings
The descriptive statistics of the participants’ judgment of severity on a four-point scale is presented in Table 4-6.
Table 4-6 The Participants’ Overall Performance on the Severity Judgment Task
Group Severity M SD F
p-value
G1 [+severity] 3.3 0.78
52.643 .000*
[-severity] 2.39 1.05
G2 [+severity] 2.93 0.76
55.675 .000*
[-severity] 2.13 0.86
G3 [+severity] 2.35 0.87
21.474 .000*
[-severity] 1.68 0.79
As Table 4-6 shows, each proficiency group of the present study performed in the same way on their severity judgment task. Significant within-group differences (G1: p = .000*, G2: p
= .000*, G3: p = .000*) were found between with or without severities through the test of One-way ANOVA, showing that the events designed in the present study were effective.
According to Brown and Levinson (1987), severity is a factor influencing the politeness strategies. Previous studies showed that severity affect the performances on different speech acts (Chang, 2005; Lin, 2010; Nureddeen, 2008; Reese-Miller, 2000; Tanck, 2002). To examine how severity affects the speech act of advice-giving, we first assessed how severity correlated to the participants’ willingness to give advice. Figure 4-4 presents percentages of the participants’ willingness6 to give advice and the mean score of their judgment of severity on a four-point scale.
[+severity] [-severity] [+severity] [-severity] [+severity] [-severity]
G1: Elementary G2: Intermediate G3: Native
Percentage of willingness 76.67 47.50 72.50 44.17 61.25 30.00
Figure 4-4 The Participants’ Willingness to Give Advice Concerning Severity
6 There were four events in the Severity Judgment Task with or without severe consequences highlighted.
Therefore, for the experimental groups, each group got the total token of responses 120 for [+severity] and [-severity] respectively, while the native controls got 80 responses. In the [+severity] condition, 92 responses in the elementary group chose to give advice under such a condition, so we got the percentage of 92/120=76.67%.
Interestingly, the elementary group rated the severity degree of events with or without severity higher than the intermediate group and the native control group did. The One-way ANOVA test showed that the performances between groups also showed significant differences (p = .000*).
Significant differences (G1: p = .000*, G2: p = .000*, G3: p = .000*) were found between the participants’ willingness to give advice and their judgment on the severity degree in each group by using the two-tailed test. As seen in Table 4-7, when the degree of severity of the event was higher, the participants were more willing to give advice. Moreover, the correlation between their willingness to give advice and the degree of severity was stronger in the control group than the correlation in the two experimental groups (G1: r = .635 < G2: r
= .636 < G3: r = .781).
Table 4-7 The Pearson’s r and Spearman’s rho of Each Group
Elementary Intermediate Native
Pearson’s r .635 .636 .781
p-value
.000* .000* .000*Spearman’s rho .614 .648 .795
p-value
.000* .000* .000*Figure 4-5 presents the descriptive statistics for the interaction between severity and the participants’ choice of advice types. The data collected other than the investigated advice types in the DCT were excluded from the table in order to be consistent with the data obtained from the MCQ. There was a significant difference in each type of advice (χ2=78.629, df =3, p = .000*).
[+severity] [-severity]
Direct-frequency 651 506
Direct-percentage 34.89 27.68
Hedged-frequency 636 558
Hedged-percentage 34.08 30.53
Indirect-frequency 326 304
Indirect-percentage 17.47 16.63
None-frequency 253 460
Figure 4-5 The Participants’ Overall Performance concerning Severity
As can be seen in Figure 4-5, the major difference between the events with or without severity lay in the choice of opting out and their preferences for direct and hedged advice.
The percentage of choosing to opt out decreased drastically in events with severe consequence highlighted. Slightly more than one-fourth of the participants (25.16%) chose to opt out in events without severe consequences highlighted; however, only 13.56% of the participants chose to opt out in events with severe consequences highlighted. As for the participants who chose to give advice, they favored direct advice the most, followed closely by hedged advice and a steep decrease in indirect advice. When giving advice in events without severe consequences highlighted, the percentage of choosing hedged advice (30.53%) was slightly higher than that of direct advice (27.68%). Likewise, there was a sharp decrease in the participants who chose indirect advice.
To be careful with the results, we examined the performance of each proficiency group
and found a significant within-group difference (p < .05), as presented in Table 4-8. Also, there was a significant within-severity difference (p < .05), as showed in Table 4-9.
Table 4-8 The p-values for the within-group Differences Concerning Types of Advice and Severity
Group χ2 df
p-value
G1: Elementary 16.569 3 0.001*
G2: Intermediate 47.440 3 .000*
G3: Native control 25.708 3 .000*
Table 4-9 The p-values for the within-severity Differences Concerning Types of Advice and Proficiency
Severity χ2 df
p-value
[+severity] 140.396 6 .000*
[-severity] 166.565 6 .000*
As showed in the above two tables, there was an interaction between severity and types of advice within each proficiency group and that there was also an interaction between proficiency level and types of advice with each severe consequences highlighted or not highlighted. The descriptive statistics of each proficiency group is presented in Figure 4-6.
[+severity] [-severity] [+severity] [-severity] [+severity] [-severity]
G1: Elementary G2: Intermediate G3: Native
Direct-frequency 276 240 259 181 116 85
Direct-percentage 39.15 35.19 36.53 26.35 25.66 18.52
Hedged-frequency 236 200 280 267 120 91
Hedged-percentage 33.48 29.33 39.49 38.86 26.55 19.83
Indirect-frequency 120 122 122 106 84 76
Indirect-percentage 17.02 17.89 17.21 15.43 18.58 16.56
None-frequency 73 120 48 133 132 207
Figure 4-6 The Performance of Each Group concerning Severity
As we can see in Figure 4-6, more participants in each group chose to opt out in events without severe consequences highlighted. The percentages of opting out of each group in events without severe consequences highlighted were higher than those in events with severe consequences highlighted. Concerning the situations when the participants wanted to give advice, generally, more direct advice was given in events with severe consequences highlighted. The elementary group performed similarly to the overall performance presented in Figure 4-5 in events with severe consequences highlighted. They chose direct advice the most, followed closely by hedged advice, and a strident decrease in direct advice. However, the intermediate group performed similarly to the overall performance presented in Figure 4-5 in events without severe consequences highlighted. Hedged advice was the most favored advice type, followed by direct advice and indirect advice.
Each proficiency group also performed significantly differently on events with or without severe consequences highlighted. First of all, the percentage of choosing to opt out got higher, as the proficiency level increased. Moreover, the disparity of opting out between
with and without severe consequences highlighted got wider as the proficiency level bettered.
Significant between-group differences were found between G1 and G2 (with severe consequences highlighted: p = 0.024*, without severe consequences highlighted: p = .000*), between G1 and G3 (with severe consequences highlighted: p = .000*, without severe consequences highlighted: p = .000*), and between G2 and G3 (with severe consequences highlighted: p = .000*, without severe consequences highlighted: p = .000*).
The native control group chose to opt out the most frequently regardless of severity;
however, the percentage of choosing to opt out was far higher in events without severe consequences highlighted (45.1%) than that in events with severe consequences highlighted (29.2%). When the native controls chose to give advice, there was no considerable disparity between direct and hedged advice. Yet, the gap between direct and indirect advice was wider in events with severe consequences highlighted.
Regarding the experimental groups, the intermediate participants chose to opt out three times more frequently in events without severe consequences highlighted (19.36%) than that in events with severe consequences highlighted (6.77%). When giving advice in events with severe consequences highlighted, though the participants still selected hedged advice the most, the gap between the percentage of choosing hedged advice and that of choosing direct advice was narrower. On the other hand, the intermediate participants made a prominent bar of hedged advice in events without severe consequences highlighted. As for the elementary group, direct advice was favored the most, followed by hedged advice in events with severe consequences highlighted. Only a small number of the participants chose indirect advice. In events without severe consequences highlighted, the elementary participants still chose direct advice the most. However, they chose direct and hedged advice less frequently than they did
in events with severe consequences highlighted, showing that more participants selected more opting-outs and indirect advice in events without consequences highlighted.
The One-way ANOVA test and the two-tailed test indicated that severity had a lot to do with the participants’ willingness to choose advice. As the degree of severity got higher, the participants were more willingly to give direct advice and less opting out. The chi-square test showed that the severity of an event influenced the participants’ choice of advice types.
Moreover, different proficiency participants had a different choice of advice in facing events with or without severe consequences highlighted.
4.3.2 General Discussion
As can be seen in Figure 4-6, the participants of each proficiency group chose more direct advice in the event with severe consequences highlighted; furthermore, fewer participants chose to opt out in the event with severe consequences highlighted. This result supports the previous research that severity has been found a factor affecting the choice of politeness strategies (Brown & Levinson, 1987; Chang, 2005; Cohen, 2005; Lin, 2010;
Nureddeen, 2008; Reese-Miller, 2000; Tanck, 2002).
There is no doubt that severity affects a person’s strategies of politeness on different speech acts (Brown & Levinson, 1987). In the speech act of apologizing, more polite apologies were employed when people encountered a more severe offense (Holmes, 1998;
Lin, 2010; Schlenker & Darby, 1981; Tsai, 2002). Brown and Levinson (1987) contended that the weightiness of an FTA boosts with the increase of offense, the relative social power, or the relative social distance. Moreover, a heavier FTA requires more elaborated or polite apologies. Additionally, it is generally predicted that more intensified Illocutionary Force
Indicating Device (IFID)7 like baoqian ‘sorry,’ and buhaoyisi ‘excuse me’ would be employed in more serious offenses. Likewise, in the speech act of advice-giving of the present study, our participants also chose more intensified IFIDs which involved the directive mood and the performative verbs in direct advice in events with severe consequences, as showed in the following examples:
(10) You shouldn’t eat hamburger, it might damage your health. (Taken from G2S13) (11) I suggest you should wear more clothes. (Taken from G2S9)
(12) I advise you go home and eat dinner. (Taken from G2S27)