2008年--外文翻譯--帶薪年假和工作時(shí)長（節(jié)選）

1、<p>　　2100單詞，1.1萬英文字符，2860漢字</p><p>　　出處：Published version: Mark Wooden and Diana Warren, ‘Paid Annual Leave and Working Hours: Evidence from the HILDA Survey’, Journal of Industrial Relations 50(4), Se

2、ptember 2008, pp. 664-670. [http://dx.doi.org/10.1177/0022185608094118] </p><p>　　Paid Annual Leave and Working Hours:</p><p>　　Evidence from the HILDA Survey Mark Wooden and Diana Warren</p&

3、gt;<p>　　University of Melbourne, Australia</p><p>　　M Wooden，D Warren</p><p><b>　　Abstract </b></p><p>　　Using data from wave 5 of the Household, Income and Labo

4、ur Dynamics in Australia (HILDA) Survey, this study examines: (i) the extent to which Australian employees use their annual leave entitlements; and (ii) the association between annual leave taking and weekly hours of wor

5、k. After restricting attention to employees likely to have entitlement to at least four weeks of paid annual leave, it is found that the mean number of days of leave taken per year is around 16 and that the majority of e

6、mplo</p><p>　　Keywords: Annual leave; Australia; HILDA Survey; Working hours </p><p>　　Introduction </p><p>　　A focal point in the recent debate about work-life balance has been the

7、 long hours many Australians work. Most of this debate has focused on the number of hours worked in a usual week, and only occasionally (e.g., Dennis, 2004) is any recognition given to the possibility that increases in t

8、he length of the working year arising from not fully utilising annual leave </p><p>　　entitlements might also be important. </p><p>　　This paper reports on the extent of usage of paid annual le

9、ave in Australia using new data collected in wave 5 of the Household, Income and Labour Dynamics in Australia (HILDA) Survey. The paper also examines the extent to which annual leave use is correlated with usual weekly

10、working hours. A priori the direction of association between these two variables is difficult to predict. Regularly working long hours should increase both the need and desire for an extended period of leave. Alternati

11、v</p><p><b>　　Data </b></p><p>　　Discussed in more detail in Wooden and Watson (2007), the HILDA Survey is a household panel survey that began, in 2001, with a large nationally repre

12、sentative sample of Australian households. All members of responding households from wave 1 form the basis of the panel to be followed over time, though interviews are only conducted with persons aged 15 years or older.

13、</p><p>　　While the survey has a longitudinal design, it employs following rules that are designed to ensure the sample maintains its cross-sectional representativeness over time. Sample composition is, how

14、ever, affected by attrition, with only 74 per cent of respondents from wave 1 still participating in wave 5. The total number of completed interviews achieved in Wave 5 was 12,759. </p><p>　　To account for

15、 possible biases, sample weights have been constructed that both correct for observable differences in the probability of different individuals responding each wave and ensure population estimates on key variables matc

16、h known totals for the population of households in Australia. The analyses in this paper make use of weighted data. </p><p>　　The data on annual leave come from responses to two questions. The first identif

17、ies whether a respondent had spent any time on paid annual leave during the 12 months preceding the interview. All persons answering in the affirmative were then asked how many days (or weeks) they spent on paid annual

18、 leave during that 12-month period. These questions were included for the first time in wave 5 and so we only have information about patterns of leave usage over a single one-year period. </p><p>　　Annual Le

19、ave Usage </p><p>　　Table 1 presents summary statistics on both the proportion of workers taking any paid annual leave and the average number of days taken by persons employed at the date of interview. This

20、 table reveals that just over half of all employed persons took at least one day of paid annual leave during the one-year reference period, with the mean number of days taken being just nine. It is, however, conventiona

21、l to ignore the self-employed (own account workers). For many of these workers the concept of</p><p>　　Restricting our attention to the employee workforce (N=6747), Table 1 confirms that a large proportion

22、 of employees do not take any paid annual leave in a one-year period, and appears to also indicate that average leave usage is only half what is generally accepted as the community standard (i.e., four weeks). The data

23、presented in Table 1, however, do not provide a good guide to how usage of leave compares with entitlements. There are at least three reasons for this. First, over one-quarter</p><p>　　A better guide to the

24、 extent to which leave entitlements are being used is provided by focusing on the sub-sample of employees who state that their employer provides them with paid annual leave, have been employed with their current employer

25、 for at least one year, and report usual weekly working hours of 35 or more (N=3242). Almost 90 per cent of this group reported taking at least one day of paid annual leave during the year, with the mean leave taken be

26、ing 16.1 days. Further information on </p><p>　　Annual Leave and Weekly Work Hours </p><p>　　We now turn to the question of whether annual leave usage is correlated with the usual number of ho

27、urs worked each week. All analyses are again restricted to the sub-group of employees who have leave entitlements, who work full-time in their main job and who, at the time of the wave 5 interview, had been employed with

28、 their current employer for at least one year. We further restrict our sample to employees working full-time hours at the time of both the wave 4 interview and wave 5 interview. </p><p>　　At first glance the

29、 data indicate that any association between weekly hours of work and annual leave usage is very small, with a Pearson correlation coefficient of just 0.04. Perhaps larger associations are being masked by intervening ass

30、ociations with other variables. Regression equations were therefore estimated that held constant a number of worker and job characteristics. </p><p>　　The dependent variable is the total number of annual le

31、ave days taken during the preceding 12 months. It is close to continuous and hence linear least squares regression methods are used. We experimented with four different specifications. The first specification restricts

32、the explanatory variables to usual weekly hours of work (in all jobs) and a number of personal characteristics. The characteristics controlled for are sex, which was interacted with both marital status and the presence

33、 of d</p><p>　　Given our interest here is in the estimated coefficients on the hours variables, only a summary of the results are presented. These can be found in Table 2. The top half of this table present

34、s the results using the‘full sample’ and show that the relationship between usual weekly hours of work and the number of paid annual leave days taken during the year is positive but relatively small in magnitude. In sp

35、ecification (1), which only includes demographic controls, the coefficient on usual hour</p><p>　　Using dummy variables instead of the continuous hours variable leads to a modification of this conclusion. I

36、t now appears that the incidence of annual leave does not rise monotonically with hours worked. Instead, days of annual leave taken peak for those reporting working 50 to 59 hours per week. Employees in this group are e

37、stimated to take 1.8 more days of leave per year than otherwise comparable workers that work a 35 to 40 hour week. Those working even longer work weeks (i.e., 60 hours or</p><p>　　The relatively high values

38、 on the Reset test, however, suggest at least three of our equations are misspecified. Experimentation with different functional forms indicates that the source of the misspecification most likely lies in outlying cases

39、 at the upper end of the distribution. These cases are distinctly different from other cases and are dominated by two distinct groups: (i) school teachers, who often report taking up to 12 weeks annual leave each year; a

40、nd (ii) employees taking a block </p><p>　　Conclusions </p><p>　　The data collected as part of wave 5 of the HILDA Survey confirm that many Australian employees do not take any paid annual leav

41、e during a one-year period and that average leave usage is only about half what is accepted as the community standard. In part, this reflects the presence of casual employees who do not have paid annual leave entitlemen

42、ts. In addition, there are sizeable groups of employees – part-time workers and workers who have been working for their current employer for less than </p><p>　　Such findings lead to the question of whether

43、 or not the taking of annual leave is inversely or positively correlated with the number of weekly hours worked. The evidence presented here suggests the relationship is positive, but small in magnitude. Slightly large

44、r effects are found when the hours variable is specified in discrete form. </p><p>　　On balance, the evidence does not suggest that persons who regularly report long hours of work each week are compensatin

45、g by taking extended periods of leave each year. Nevertheless, the data also provide no evidence to support the hypothesis that the pressures at work that might lead many people to regularly work very long hours each we

46、ek also cause them to forego their annual leave entitlements. </p><p>　　帶薪年假和工作時(shí)長：根據(jù)澳大利亞墨爾本馬克.伍登與戴安娜.華倫大學(xué)的希爾達(dá)調(diào)查證據(jù)分析</p><p>　　摘要：根據(jù)澳大利亞希爾達(dá)調(diào)查的第五次家庭、收入與勞動(dòng)動(dòng)態(tài)數(shù)據(jù)，研究檢驗(yàn)了（1）澳大利亞人如何最大程度使用帶薪年假權(quán)利；（2）年假與每周工

47、作時(shí)長的關(guān)系。將范圍縮小至帶薪年假至少4周的員工，結(jié)果發(fā)現(xiàn)實(shí)際帶薪年假為16天左右，且多數(shù)員工（63%）年假少于20 天。此事件表明帶薪年假與周工作時(shí)長呈正相關(guān)，但其影響力度卻極其微小。</p><p>　　關(guān)鍵詞：年假；澳大利亞；希爾達(dá)調(diào)查；工作時(shí)長</p><p><b>　　引言</b></p><p>　　近期關(guān)于工作與生活如何平衡的爭

48、論聚焦于澳大利亞人的工作時(shí)間過長。話題多為每周工作時(shí)長，少數(shù)情況人們認(rèn)同工作時(shí)間增加是由未充分利用年假導(dǎo)致的，這一點(diǎn)很重要。</p><p>　　根據(jù)澳大利亞（希爾達(dá)）調(diào)查搜集的第五次家庭、收入和勞動(dòng)動(dòng)態(tài)的新數(shù)據(jù)，本文報(bào)告了澳大利亞人帶薪年假的休假情況，同樣檢驗(yàn)了帶薪年假與周工作時(shí)長的相關(guān)性，此前研究表明而這兩種變量的關(guān)系難以預(yù)測(cè)。通常，工作時(shí)間越長，員工延長休假的需要與愿望便越強(qiáng)。此外，長時(shí)工作反映了工作時(shí)間壓

49、力，這一點(diǎn)可能阻礙員工應(yīng)有的休假。</p><p><b>　　數(shù)據(jù)</b></p><p>　　在伍登與華生（2007）中的描述更為詳細(xì)，希爾達(dá)調(diào)查是家庭專項(xiàng)調(diào)查，開始于2001年，由居住在澳大利亞眾多來自各國的家庭代表組成。雖然只對(duì)15歲左右的成員進(jìn)行了采訪，但這第一組家庭成員代表形成了這一專項(xiàng)調(diào)查的基礎(chǔ)，此后的調(diào)查都以此為樣本。</p><p

50、>　　研究采用了縱向設(shè)計(jì)，運(yùn)用一些規(guī)則以確保樣本在長時(shí)間內(nèi)維持其代表的典型性。然而，樣本結(jié)構(gòu)卻受到成員不足的影響，第一組代表中僅有75%參加了第五次調(diào)查，五次調(diào)查的總采訪次數(shù)為12759次。</p><p>　　考慮到可能出現(xiàn)的偏差，樣本權(quán)重構(gòu)成既符合觀察每一次調(diào)查中個(gè)體的多樣性，又保證對(duì)關(guān)鍵人數(shù)變量的估計(jì)符合澳大利亞中體家庭人數(shù)。本文分析使用了權(quán)重?cái)?shù)據(jù)。</p><p>　　關(guān)于帶

51、薪年假的數(shù)據(jù)來源于兩個(gè)問題的回答。第一個(gè)問題明確了員工是否在采訪前12個(gè)月中已帶薪休假。休過假的員工需回答他們帶薪休假的天數(shù)。這些問題首次收入在第五次調(diào)查中，因此我們僅有一年的帶薪休假信息模版。</p><p><b>　　年假利用情況</b></p><p>　　表1 展示了關(guān)于員工帶薪年假的比例和以及受訪者的平均年假天數(shù)的數(shù)據(jù)總結(jié)，表格顯示在可供參考的年限內(nèi)，在有

52、九天年假的員工中，僅過半員工享有多于一天的帶薪休假。然而，通常個(gè)體經(jīng)營者不包括在內(nèi)。對(duì)他們而言，不工作就意味著無收入，帶薪休假對(duì)他們來說毫無意義。這一說法表明股份制企業(yè)家同樣不包括在內(nèi)。</p><p>　　將研究范圍縮小至雇員勞動(dòng)力（N=6747），表1 證實(shí)了多數(shù)員工未有任何帶薪休假，且似乎同樣表明帶薪休假作為社區(qū)標(biāo)準(zhǔn)（如：4周）僅得到半數(shù)認(rèn)可。然而，表1所展示的數(shù)據(jù)未為帶薪休假作為一項(xiàng)權(quán)利提供好的指導(dǎo)，其有

53、三個(gè)原因：首先，四分之一多的勞動(dòng)力為流動(dòng)勞動(dòng)力，因?yàn)椴幌碛袔侥昙伲黄浯?，兼職勞?dòng)力的帶薪年假按比例分配，因此，長期兼職工作者的帶薪年假少于4周；第三，表1所囊括的一些員工雇傭期少于一年，因此沒有帶薪年假。</p><p>　　關(guān)注員工子樣本一定程度上為帶薪休假權(quán)利提供了更好的指導(dǎo)，這些員工表示公司領(lǐng)導(dǎo)準(zhǔn)許帶薪休假，且他們的工齡至少有一年，每周工作35個(gè)小時(shí)以上（N=3242）。這一組中90%以上的人年假有16.

54、1天的，至少一天可帶薪休假，。圖1提供了此帶薪休假模版更多信息，表明了此類假期分布的廣泛性。員工假期普遍為20天（4周），而多數(shù)人（63%）年假少于20天。</p><p><b>　　年假與每周工作時(shí)長</b></p><p>　　現(xiàn)在問題轉(zhuǎn)向年假是否與每周工作時(shí)長相關(guān)。所有分析僅限于有帶薪年假的全職員工。在第五次采訪中,他們的工齡至少一年。在第四次和第五次采訪中，

55、我們將員工范圍縮小至全職員工。</p><p>　　這些數(shù)據(jù)首先表明，每周工作時(shí)長和帶薪休假的關(guān)系不大，皮爾森關(guān)系指數(shù)顯示為0.04?；蛟S其他變量與此關(guān)系更大。因此研究使用回歸公式估算員工和工作特點(diǎn)的常數(shù)。</p><p>　　因變量是前一年內(nèi)帶薪休假天數(shù)的總和。這一變量幾乎是連續(xù)不變的，因此運(yùn)用了最少線性平方回歸原理。我們對(duì)四個(gè)不同的規(guī)格進(jìn)行了實(shí)驗(yàn)。第一個(gè)規(guī)格將說明變量局限于周工作時(shí)長（

56、所有工作中）和一系列個(gè)別特點(diǎn)中。受控制的特點(diǎn)有性別，此特點(diǎn)與性別、兒童（7個(gè)虛擬變量）、年齡（4個(gè)虛擬變量）、國籍、出生地（2個(gè)虛擬變量表明海外出生以及區(qū)別出生于英語國家和其他國家的人）和居住地點(diǎn)（3個(gè)虛擬變量基于居住地與主城市的距離以區(qū)別受訪者）相互作用。第二個(gè)規(guī)格管理工齡（連續(xù)計(jì)算工齡年限并在第四次采訪中以虛擬變量辨別工齡少于一年的參與者）、職位（9個(gè)虛擬變量）、定期合同和自定義自由職業(yè)（分析中不包括自由職業(yè)）、勞動(dòng)力工作者、公共事

57、業(yè)職業(yè)、聯(lián)盟成員和公司規(guī)模（2個(gè)虛擬變量）。第三個(gè)規(guī)格包括計(jì)算員工一年中帶薪假期和無薪假期的時(shí)長。對(duì)于第四個(gè)規(guī)格，我們測(cè)試了周工作時(shí)長和帶薪假期非線性聯(lián)系的可能性，并用3個(gè)虛擬變量（區(qū)別周工作時(shí)長分別為41-49小時(shí)、50-59小時(shí)以及60小時(shí)以上的員工，工作時(shí)長為35-40小時(shí)的員工為一個(gè)控制組）代替了連續(xù)時(shí)長變量。除了計(jì)算假期時(shí)長，所有控制變量都在第四次采訪中使用。</p><p>　　若在預(yù)計(jì)系數(shù)中關(guān)系重大

58、的是小時(shí)變量，那么展現(xiàn)的便僅是對(duì)結(jié)果的總結(jié)（表2）。表2上半部分是全樣本的結(jié)果，表示周工作時(shí)長和帶薪休假天數(shù)的正相關(guān)關(guān)系，但其關(guān)系相關(guān)幅度較小。規(guī)格（1）中僅包括人數(shù)控制，而一般時(shí)長系數(shù)數(shù)據(jù)尤為重要，表明每周額外增加10小時(shí)工作與當(dāng)年額外帶薪假期有關(guān)。然而，增加工作控制將產(chǎn)生很大不同，不僅模型的說服力顯著增加，一般時(shí)長變量系數(shù)也將增長過半并不會(huì)對(duì)傳統(tǒng)95%可信度上產(chǎn)生影響。這一發(fā)現(xiàn)在使用變量計(jì)算其他類型帶薪假期中，有同樣效果。</

59、p><p>　　以虛擬變量替代連續(xù)小時(shí)變量，使結(jié)論得以改進(jìn)，即年假不隨工作時(shí)長增加而直線遞增，而是年假時(shí)間在周工作時(shí)長為50-59小時(shí)達(dá)到峰值。與其他每周工作35-40小時(shí)員工相比，此組員工年假估計(jì)平均多出1.8天。而工作時(shí)間更長的小組（60小時(shí)及以上）的年假則不比每周工作35-40小時(shí)的員工長。</p><p>　　然而在二次測(cè)試中，得到的相對(duì)高值表明至少三個(gè)等式的規(guī)格錯(cuò)誤。不同功能形式的實(shí)

60、驗(yàn)表示規(guī)格錯(cuò)誤可能在于頂層案例分布不當(dāng)。這些案例與其他案例不同，是因?yàn)橛袃蓚€(gè)顯著的主要小組：（1）學(xué)校教師（通常年假為12周）；（2）公司員工（假期總和）。因此，在排除年假50天以上的情況，我們?cè)俅喂烙?jì)了等式規(guī)格。結(jié)果總結(jié)于表3下半部分。在這些規(guī)格中，再測(cè)試得出的數(shù)據(jù)都很小且作用不大，這表示錯(cuò)誤規(guī)格不是出現(xiàn)偏差的源頭。更為重要的是，小時(shí)變量得出的結(jié)果差別微小，因此我們得出的結(jié)論未有任何影響。</p><p>&l

61、t;b>　　結(jié)論</b></p><p>　　第五次希爾達(dá)調(diào)查中收集的數(shù)據(jù)證實(shí)了許多澳大利亞人沒有帶薪年假且平均年假僅是社區(qū)標(biāo)準(zhǔn)的一半。這一結(jié)論部分反映了沒有帶薪年假的自由職員出現(xiàn)的原因。此外，還有不同大小組的職員——兼職工作者和工齡少于一年的職員（于他們而言，不享有4周年假）。然而，對(duì)于年假至少4周的職工而言，他們的實(shí)際假期僅有16天，且多數(shù)人（63%）年假少于20天。</p>