| 查看: 718 | 回复: 0 | ||
[求助]
论文投稿，主编关于统计的意见我不明白，请帮我看一看。
|
|
主编的第一次意见 Table 2 - there is no description of correlation having been undertaken. Further to this the use of simple correlation analysis is fraught with danger and the authors should consult Hassleer & Thadewald (2003) - The Statistician 52(3) 367-379 for detail on multivariate linear modelling. 我们的修改如下： Statistical analyses -------------------------------。Pearson's correlation coefficients were computed to quantify linear associations between physicochemical and structural parameters of MP. Table 2. Pearson correlation coefficient between sodium nitrite and physicochemical and structural parameters of MP. CA FA SH S-S HY TU SO 3-NT TABRS FA -0.636** SH 0.815** -0.623** S-S -0.771** 0.584** -0.967** HY 0.768** -0.508* 0.699** -0.759** TU 0.637** -0.184 0.539* -0.538* 0.492* SO 0.525* -0.259 0.486* -0.557* 0.719** 0.495* 3-NT -0.425 0.334 -0.557* 0.468* -0.152 -0.397 -0.138 TABRS 0.620** -0.532* 0.674** -0.581** 0.373 0.756** 0.293 -0.753** SN -0.636** 0.628** -0.790** 0.733** -0.489* -0.492* -0.214 0.812** -0.829** Note: *means significant difference (P < 0.05), **means significant difference (P < 0.01). Abbreviations in the table: CA, carbonyl; FA, free amine; SH, sulfydryl; S-S, disulfide; HY, hydrophobicity; TU, turbidity; SO, solubility; 3-NT, 3-nitrotyrosine; SN, sodium nitrite. 我们实在不明白主编关于相关分析的意思，他是想我们使用多元线性模型分析Table2的参数之间的相关性吗？于是就给了主编写信咨询，主编的第二次回信是这样的： 第二次去信 Dear -------, Thank you for your help to improve our manuscript. In your last letter, you mentioned ‘correlation analysis in table 2’. We are sorry for our unclear description. Pearson's correlation coefficients have been computed to quantify linear associations between physicochemical and structural parameters of MP. We send table 2 to you to see if we understand you. Again, thank you for taking the time and energy to help us improve the paper. We will do further modifications if you think we should make more modifications on table 2. 主编的第二次回信： Anyway the point I made was that in the use of correlation analysis you need to be very careful. This only tells the level of association, not causation and to overcome this as pointed out in the reference I sent you should undertake a GLM analysis that accounts for the variance from the treatment structure which will indicate the correlation between traits once the structure is accounted for. Please also note that if you applied replicates then this needs to be taken account of in the analysis. 请问主编到底是什么意思，多元线性模型分析参数之间的相关性不知道该如何进行？此外主编第二次回信中的‘treatment structure’是什么意思？谢谢！ |
回复此楼» 猜你喜欢
各位大神，目前国内有哪些比较好用的逆合成软件？
已经有13人回复
-
国社科系统bug了，是不是要放榜了？
已经有10人回复
-
现代”学阀”该如何界定
已经有9人回复
-
课题组招2027级博士 上海工程技术大学 激光智能制造方向
已经有4人回复
-
上海工程技术大学 激光智能制造课题组 2027级博士研究生招生
已经有3人回复
-
上海工程技术大学激光智能制造课题组｜2027级博士研究生招生公告
已经有11人回复
-
上海工程技术大学激光智能制造课题组-博士研究生
已经有10人回复
-
申博发邮件
已经有11人回复
-
我的奶奶
已经有3人回复