| 查看: 423 | 回复: 1 | ||
| 【悬赏金币】回答本帖问题，作者jolil将赠送您 5 个金币 | ||
[求助]
求助回复意见 已有1人参与
|
||
|
One major issue I have with this paper is why the authors used the Duncan's multiple range test for statistical analysis. This statistical method is particularly prone to type I errors. I suggest re-doing the analysis with a more conservative analysis like ANOVA followed by a simple student t-test or another method. Probably related to this, many of the figures have differences between groups that don't look significantly different to me. For example, in Figure 1B are G vs P groups really different? In Figure 5, are NG vs P or P vs MG really different? There are many more examples, but I would like to know what the interpretation is like with a more conservative statistical analysis. I'm not a statistician so I wouldn't reject the paper even if another analysis comes to a different conclusion, but at the minimum the authors should perform the analysis with another method and mention the differences (if any) that occur. 这个审稿人提出用ANOVA followed by a simple student t-test 这个统计方法，请问有这种方法吗？可是我是多组两两比较，能用t-test吗 |
回复此楼» 猜你喜欢
求材料调剂
已经有7人回复
-
一志愿天大材料与化工（085600）总分338
已经有3人回复
-
085600材料与化工
已经有5人回复
-
085600材料与化工调剂 324分
已经有6人回复
-
286求调剂
已经有9人回复
-
焦虑
已经有12人回复
-
344求调剂
已经有6人回复
-
266求调剂
已经有9人回复
-
化学工程321分求调剂
已经有18人回复
-
314求调剂
已经有8人回复
hujun
金虫 (小有名气)
- 应助: 0 (幼儿园)
- 金币: 711.2
- 散金: 20
- 帖子: 221
- 在线: 47.2小时
- 虫号: 561660
- 注册: 2008-05-22
- 性别: MM
- 专业: 法医毒理、病理及毒物分析
2楼2017-03-10 12:40:17
