作者apson (小李)
看板Statistics
标题Re: [问题] 关於李克特尺度
时间Tue Aug 18 16:08:46 2009
请教先进有关李克特尺度的问题,最近有一篇期刊的评审问了我一个问题
我在文章里资料分析段写道“Before statistic analysis, tests were completed for
normality, homoscedasticity, linearity and independence of error terms
(Cohen& Cohen, 1983). These assumptions were examined and were found to hold
true.”
评审问:I don’t understand the meaning of this explanation. The variables are
5-point Likert scale. Why do they need to be normal, homoscedastic, and
linear? Even weirder thing is that these assumptions were found to hold true.
Do the authors mean the error terms of the multiple regression model? If so,
it shouldn’t come as the first paragraph of the discussion part (even before
the factor analysis part).
请问Likert scale的问项在进行回归或因素分析之前是否要做这些检定呢?因为Hair教科
书里很多分析之前都要先分析是否符合假设,但是没针对尺度特别说明一定要或一定不要
做这些分析
这让我蛮困扰的,不知道我有没有会错评审的意思?评审是指Likert scale的尺度特性本
来就不必做这些测试吗?
--
※ 发信站: 批踢踢实业坊(ptt.cc)
◆ From: 122.127.176.224
※ apson:转录至看板 PhD 08/18 17:52
1F：→ bmka:The 5-point Likert scale apparently has a discrete 08/18 20:47
2F：→ bmka:distribution, but you performed a test to check 08/18 20:49
3F：→ bmka:whether it has a normal distirbution, which is used 08/18 20:51
4F：→ bmka:to model the distribution of continuous variables. 08/18 20:51
5F：→ dswchen:statistic analysis? or statistical analysis? 08/18 22:21
6F：→ dswchen:before statistic analysis?原来你那些test都不是统计分析 08/18 22:22
7F：→ dswchen:其实我一直觉得 台湾那些作问卷跟评量的 都像你一样乱搞 08/18 22:23
8F：推 wlsherica:楼上第三句话有点太重,并没有"都"好吗? 08/18 22:31
9F：→ takanaka:没有见到review全文也不知道原作的领域为何 08/18 22:56
10F：→ takanaka:但推测reviewer的意思是Likert scale有其本来的限制 08/18 22:57
11F：→ takanaka:比方说很难成立这个rating是linear,等距的 08/18 22:59
12F：→ takanaka:社会科学研究实务常是在工具限制下尽量做推论 08/18 23:00
13F：→ takanaka:因此reviewer或许觉得硬把一个不完美的工具说成是完美的 08/18 23:03
14F：→ takanaka:是有点不合 08/18 23:03
15F：→ takanaka:另，社会科学的兴趣在统计背後的假设，而非统计本身 08/18 23:05
16F：→ takanaka:如果因为工具限制或资料分布不理想就不做了 08/18 23:05
17F：→ takanaka:似乎是因噎废食。质性方法、量化方法并存的意义我想在这 08/18 23:06
18F：→ apson:我接受bmka的说法,但是实际上likert scale在社科研究里常用 08/19 00:38
19F：→ apson:当然likert scale有其限制 08/19 00:39
20F：→ bmka:There is nothing wrong with fitting a regression model 08/19 00:41
21F：→ bmka:to Likert scale data, if the sample size is large enough. 08/19 00:42
22F：→ bmka:Review算是有点挑语病, 但是作者也要对自己写的东西负责 08/19 00:43
23F：→ bmka:Reviewer 08/19 00:44
24F：→ apson:我知道要负责&sample size够大 08/19 00:45
25F：→ bmka:因为那段文字读起来的感觉是作者只是follow其他人做的分析 08/19 00:46
26F：→ bmka:却不是真的懂自己用了什麽统计方法 08/19 00:46
27F：→ apson:受教了,谢谢 08/19 00:47
28F：→ bmka:写paper有时这些小细节要小心 :) 08/19 00:49