nqkk58 发表于 2024-9-30 01:25:02

文案中统计学办法的描述


    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">统计学<span style="color: black;">办法</span><span style="color: black;">始终</span>是在科技期刊<span style="color: black;">科研</span>原著<span style="color: black;">文案</span>中的重要部分，在<span style="color: black;">科研</span>设计和样本量估算完成的<span style="color: black;">状况</span>下，最重要的<span style="color: black;">便是</span>数据的统计学分析环节，但<span style="color: black;">日前</span>中国大<span style="color: black;">都数</span>医学期刊对统计学<span style="color: black;">办法</span>描述的<span style="color: black;">需求</span>还<span style="color: black;">不足</span>规范。现将统计学<span style="color: black;">办法</span>的描述要点作一整理，作者撰写<span style="color: black;">科研</span>原著类<span style="color: black;">文案</span>时，应遵循以下<span style="color: black;">过程</span>描述统计学<span style="color: black;">办法</span>。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">1 描述统计学软件信息：</span></strong><span style="color: black;">首要</span>要对正文的统计学软件加以描述，需描述软件名<span style="color: black;">叫作</span>、<span style="color: black;">源自</span>厂家和版本。常用4大软件<span style="color: black;">包含</span>SAS、STATA、SPSS软件等，还<span style="color: black;">包含</span>附带统计学功能的软件GraphPad Prism 7软件等。</span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">2描述统计学指标：</span></strong>需说明<span style="color: black;">科研</span>中各描述性结果指标的<span style="color: black;">暗示</span>方式。</span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">2.1计量资料：</span></strong><span style="color: black;">正态分布资料<span style="color: black;">包含</span>均值(mean)，标准差(SD)，标准误(SEM)；非正态分布资料：中位数Median(M)，四分位数间距(P75-P25)。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">2.2计数资料和等级资料：</span></strong><span style="color: black;"><span style="color: black;">重点</span>为<span style="color: black;">形成</span>比(如4/15，比重)和百分率(如63%，频率强度)。<span style="color: black;">通常</span>用n(%)<span style="color: black;">暗示</span>，如想<span style="color: black;">认识</span><span style="color: black;">各样</span><span style="color: black;">疾患</span>在特定人群所占的比例，用<span style="color: black;">形成</span>比。如想<span style="color: black;">认识</span>哪一个年龄组发病率高，用百分率<span style="color: black;">暗示</span>。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">2.3效应量指标：</span></strong><span style="color: black;"><span style="color: black;">重点</span><span style="color: black;">包含</span>比值比(odds ratio，OR)，相对危险度(Risk Ratio，RR)，95%CI。如OR(95%CI)=2.6(1.3-5.2)。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3统计学分析<span style="color: black;">办法</span>：</span></strong></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.1计量资料的比较</span></strong></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.1.1两组比较(正态分布)：</span></strong></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">t检验：</span></strong><span style="color: black;">又<span style="color: black;">叫作</span>Student t检验，必须满足正态性，方差齐<span style="color: black;">要求</span>，<span style="color: black;">重点</span><span style="color: black;">包含</span>两样本t检验(独立样本t检验，成组t检验)和配对样本t检验。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">配对样本t检验的适用<span style="color: black;">状况</span>：</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">(1) 同一<span style="color: black;">科研</span>对象给予处理前、后比较(即<span style="color: black;">自己</span>配对)；</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">(2) 同一受试对象接受两种<span style="color: black;">区别</span>的处理；</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">(3) 配对的两个受试对象分别接受两种<span style="color: black;">区别</span>的处理；</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">(4) 同一对象的两个部位给予<span style="color: black;">区别</span>的处理。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">3.1.2多组比较(正态分布)：</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">方差分析：两个及以上样本间均数的比较，采用成组和配伍设计，<span style="color: black;">包含</span>单<span style="color: black;">原因</span>、双<span style="color: black;">原因</span>、多<span style="color: black;">原因</span>、析因设计、重复<span style="color: black;">测绘</span>方差分析(＞3个时间点数据比较)。在双<span style="color: black;">原因</span>、多<span style="color: black;">原因</span>、析因设计方差分析结果中<span style="color: black;">必定</span>要有主效应和交互效应的说明。<span style="color: black;">重视</span>，比较多组数据时，<span style="color: black;">不可</span>用t检验代替方差分析：<span style="color: black;">重点</span><span style="color: black;">原由</span>是t检验破坏了原先的整体设计；<span style="color: black;">显现</span>假阳性错误的概率<span style="color: black;">明显</span><span style="color: black;">增多</span>；t检验割裂了各<span style="color: black;">原因</span>之间的内在联系，<span style="color: black;">没法</span>考察交互<span style="color: black;">功效</span><span style="color: black;">是不是</span><span style="color: black;">拥有</span><span style="color: black;">明显</span>性<span style="color: black;">道理</span>。应采用方差分析结合事后检验进行两两比较。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.1.3非正态分布计量数据的非参数检验(秩和检验)：</span></strong><span style="color: black;">两组差异比较用Mann-Whitney U检验，多组差异比较用Kruskal-Wallis H法(H-检验)。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.2 计数资料的比较</span></strong></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.2.1两组比较：</span></strong><span style="color: black;">行χ2检验。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">n&gt;40并且<span style="color: black;">因此</span>理论数(T)大于5，则用Pearsonχ2检验；</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">n&gt;40并且<span style="color: black;">因此</span>理论数(T)大于1并且<span style="color: black;">最少</span>存在一个理论数&lt;5，则用校正Pearsonχ2检验；</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">n&gt;40或存在理论数(T)&lt;1，则用精确(Fisher)概率法；</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">n&lt;40：用精确(Fisher)概率法。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">配对样本资料比较：可用配对四格表χ2检验；<span style="color: black;">自己</span>前后比较：McNemyerχ2检验。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.2.2多组比较：</span></strong><span style="color: black;">行列表χ2检验。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.3等级资料的比较：</span></strong><span style="color: black;">对等级资料的<span style="color: black;">明显</span>性检验采用非参数检验法。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.3.1两组比较：</span></strong><span style="color: black;">成组设计资料用Wilcoxon两样本比较法，配对设计资料用符号秩和检验法。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.3.2多组比较：</span></strong><span style="color: black;">成组设计用Kruskal-Wallis H法(H-检验)、Ridit法；多个样本两两比较用Nemenyi法；配伍组设计用Friedman秩和检验法。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.4 <span style="color: black;">关联</span>和回归分析</span></strong></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.4.1<span style="color: black;">关联</span>性分析：</span></strong><span style="color: black;">先作散点图，确定有线性趋势方可进行<span style="color: black;">关联</span>性分析。线性<span style="color: black;">关联</span>：Pearson<span style="color: black;">关联</span>性分析(正态分布)；秩<span style="color: black;">关联</span>：Spearman<span style="color: black;">关联</span>性分析(非正态分布，等级资料)</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.4.2线性回归：</span></strong><span style="color: black;"><span style="color: black;">包含</span>因变量(<span style="color: black;">结果</span>)、自变量(<span style="color: black;">原因</span>) 和连续变量，数据需符合正态分布。简单线性回归：1个因变量，1自变量；多重线性回归：1个因变量，多个自变量。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.4.3 Logistics回归：</span></strong><span style="color: black;"><span style="color: black;">包含</span>因变量(<span style="color: black;">结果</span>)和自变量(<span style="color: black;">原因</span>)。<span style="color: black;">要求</span>Logistics回归(配对，病例对照数据)，非<span style="color: black;">要求</span>Logistics回归(成组数据)。其中非<span style="color: black;">要求</span>Logistics回归<span style="color: black;">包含</span>2种，二元Logistic回归：<span style="color: black;">指的是</span>因变量为二<span style="color: black;">归类</span>变量(是，否；患病，未患病)的回归分析；多元Logistic回归：<span style="color: black;">指的是</span>因变量为有序或无序<span style="color: black;">归类</span>变量(轻、中、重；高中、低；优、良、中、差；A，B，C，D)的回归分析。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.4.4 Cox回归：</span></strong><span style="color: black;"><span style="color: black;">包含</span>因变量(<span style="color: black;">结果</span>)和自变量(<span style="color: black;">原因</span>)，多用于<span style="color: black;">存活</span>分析。<span style="color: black;">危害</span>函数比(hazard ratio，HR)：是<span style="color: black;">存活</span>分析资料中用于估计<span style="color: black;">由于</span>某种<span style="color: black;">原因</span>的存在而使死亡/缓解/复发等<span style="color: black;">危害</span>改变的倍数。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">3.4.5纳入回归模型的变量<span style="color: black;">选取</span>：</span></strong><span style="color: black;">单<span style="color: black;">原因</span>分析后，应当<span style="color: black;">思虑</span>应该将<span style="color: black;">那些</span>自变量纳入回归模型进行多<span style="color: black;">原因</span>分析？<span style="color: black;">通常</span><span style="color: black;">状况</span>下，<span style="color: black;">意见</span>纳入的变量有：</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">(1)单<span style="color: black;">原因</span>分析差异有<span style="color: black;">明显</span>性<span style="color: black;">道理</span>的变量(此时，最好将P值放宽<span style="color: black;">有些</span>，<span style="color: black;">例如</span>0.1或0.15等，避免漏掉<span style="color: black;">有些</span>重要<span style="color: black;">原因</span>)；</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">(2)单<span style="color: black;">原因</span>分析时，<span style="color: black;">无</span><span style="color: black;">发掘</span>差异有<span style="color: black;">明显</span>性<span style="color: black;">道理</span>，<span style="color: black;">然则</span>临床上认为与因变量关系密切的自变量。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">4 统计学<span style="color: black;">办法</span>描述举例：</span></strong></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">4.1数据描述：</span></strong><span style="color: black;">①<span style="color: black;">实验</span>采用SPSS 22.0软件(美国IBM<span style="color: black;">机构</span>)和SAS 9.2软件(美国SAS Institute Inc.<span style="color: black;">机构</span>)进行分析。②统计分析时先<span style="color: black;">检测</span>各<span style="color: black;">科研</span>中心完成例数、病例的脱落<span style="color: black;">状况</span>，<span style="color: black;">而后</span>进行两组<span style="color: black;">病人</span>入选时的人口统计学及基线各<span style="color: black;">相关</span>特征的分析，考察<span style="color: black;">实验</span>组和对照组之间的可比性。③计量资料采用均数、标准差、中位数、最小值和最大值、上下四分位数进行统计描述，计数资料采用例数和百分比进行统计描述。</span></span></p>
    <p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><strong style="color: blue;"><span style="color: black;">4.2统计分析<span style="color: black;">办法</span>：</span></strong><span style="color: black;">①<span style="color: black;">实验</span>中各随访时间点两组间L2-4、股骨颈、Ward’s三角区骨密度值较基线的差值、血清钙、甲状旁腺素、骨钙素、白细胞介素<span style="color: black;">十、</span>白细胞介素6、肿瘤坏死因子α和胰岛素样生长因子1水平比较采用两样本t检验(数据正态分布)或Mann-Whitney U检验(数据非正态分布)。②组内各时间点<span style="color: black;">以上</span>数据比较采用重复<span style="color: black;">测绘</span>方差分析及LSD检验比较。③两组不良反应<span style="color: black;">出现</span>率的比较采用Pearson χ2检验。④各组骨密度值指标、骨质疏松指标及炎性因子指标间的<span style="color: black;">关联</span>性分析采用Pearson<span style="color: black;">关联</span>分析法(数据正态分布)或Spearman<span style="color: black;">关联</span>分析法(数据非正态分布)。⑤检验水准(双侧)α = 0.05。</span></span></p>
    <div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-qvj2lq49k0/65e178ae63db419d93f7bbaa0bfc5428~noop.image?_iz=58558&amp;from=article.pc_detail&amp;lk3s=953192f4&amp;x-expires=1727799684&amp;x-signature=gudYBtZa2rBCp5SlaRYwc%2BY0JsY%3D" style="width: 50%; margin-bottom: 20px;"></div>
    <div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-qvj2lq49k0/7ffeadf5b4124457931c26395459a7f7~noop.image?_iz=58558&amp;from=article.pc_detail&amp;lk3s=953192f4&amp;x-expires=1727799684&amp;x-signature=LGG%2Bum9%2Bwm1tnKSJZqKGyzooKWM%3D" style="width: 50%; margin-bottom: 20px;"></div>




j8typz 发表于 2024-10-7 16:52:13

外贸论坛是我们的，责任是我们的，荣誉是我们的，成就是我们的，辉煌是我们的。

qzmjef 发表于 2024-11-13 16:22:46

大势所趋，用于讽刺一些制作目的就是为了跟风玩梗，博取眼球的作品。

wrjc1hod 发表于 前天 21:30

你的话语如春风拂面，温暖了我的心房，真的很感谢。
页: [1]
查看完整版本: 文案中统计学办法的描述