<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0">
   <channel>
      <title>车东[Blog^2]</title>
      <link>http://www.chedong.com/blog/</link>
      <description>良好引用，良好结构，良好导航 Well referenced and well organized, with easy navigation</description>
      <language>zh-cn</language>
      <copyright>Copyright 2012</copyright>
      <lastBuildDate>Fri, 07 Jan 2011 18:28:05 +0800</lastBuildDate>
      <generator>http://www.sixapart.com/movabletype/</generator>
      <docs>http://blogs.law.harvard.edu/tech/rss</docs> 

            <item>
         <title>电脑屏幕分辨率分布统计（基于CNZZ统计数字）</title>
         <description><![CDATA[<p><a href="http://data.cnzz.com/main.php?s=resolve"><img border="0" src="http://www.chedong.com/blog/archives/2011/01/07/2011-01-07_181715.png" /></a></p><ul><li>1024x已经低于40%；</li><li>疑问： 1366x 增加最多？</li><li>疑问： 800x 略有增加？</li><li>2010年12月份的其他分辨率的量为0%， 2009年还有5.9%；<br /></li></ul><p>按宽度汇总的分布： 统计如下 <br /></p><p>&nbsp;</p><table width="351" cellspacing="0" cellpadding="0" border="0">      <tbody><tr style="height: 22px">   <td align="center" class="xl65" style="height: 22px; width: 94px">分辨率</td>   <td align="center" class="xl66" style="width: 99px">2009年12月</td>   <td align="center" class="xl66" style="width: 99px">2010年12月</td>   <td align="center" class="xl65" style="width: 59px"><strong>变化量</strong></td>  </tr>  <tr style="height: 22px">   <td align="center" class="xl65" style="height: 22px">1024x</td>   <td align="center" class="xl67">46.5%</td>   <td align="center" class="xl67">39.3%</td>   <td align="center" class="xl67"><strong>-7.2%</strong></td>  </tr>  <tr style="height: 22px">   <td align="center" class="xl65" style="height: 22px">1280x</td>   <td align="center" class="xl68">21.4%</td>   <td align="center" class="xl68">23.7%</td>   <td align="center" class="xl68"><strong>2.3%</strong></td>  </tr>  <tr style="height: 22px">   <td align="center" class="xl65" style="height: 22px">1440x</td>   <td align="center" class="xl67">15.2%</td>   <td align="center" class="xl67">18.3%</td>   <td align="center" class="xl67"><strong>3.1%</strong></td>  </tr>  <tr style="height: 22px">   <td align="center" class="xl65" style="height: 22px">1366x</td>   <td align="center" class="xl67">2.6%</td>   <td align="center" class="xl67">9.1%</td>   <td align="center" class="xl67"><strong>6.4%</strong></td>  </tr>  <tr style="height: 22px">   <td align="center" class="xl65" style="height: 22px">1680x</td>   <td align="center" class="xl67">3.5%</td>   <td align="center" class="xl67">4.4%</td>   <td align="center" class="xl67"><strong>1.0%</strong></td>  </tr>  <tr style="height: 22px">   <td align="center" class="xl65" style="height: 22px">1152x</td>   <td align="center" class="xl67">3.1%</td>   <td align="center" class="xl67">3.2%</td>   <td align="center" class="xl67"><strong>0.1%</strong></td>  </tr>  <tr style="height: 22px">   <td align="center" class="xl65" style="height: 22px">800x</td>   <td align="center" class="xl67">1.9%</td>   <td align="center" class="xl67">2.1%</td>   <td align="center" class="xl67"><strong>0.2%</strong></td>  </tr> </tbody></table><p>&nbsp;</p>]]></description>
         <link>http://www.chedong.com/blog/archives/001486.html</link>
         <guid>http://www.chedong.com/blog/archives/001486.html</guid>
                  <category domain="http://www.sixapart.com/ns/types#category">日志分析|Analytics</category>
        
        
         <pubDate>Fri, 07 Jan 2011 18:28:05 +0800</pubDate>
      </item>
            <item>
         <title>又到一年校招时： 校园用户使用的招聘类网站对比</title>
         <description><![CDATA[<p>最近<a href="http://job.sogou.com/">搜狗</a>在进行校园招聘，面试过程中对北京IT类应届生对招聘类网站的需求也做了一些调查。 应届生中的典型使用情景是这样的：</p>

<p>搜集：<a href="http://bbs.yingjiesheng.com/">应届生</a>网站搜集各种面试经验/消息；<br />
订阅：利用<a href="http://www.guolairen.com/">过来人</a>订阅感兴趣的公司面试消息（手机订阅）；<br />
提交：在各种招聘类网站中<a href="http://www.chinahr.com/">中华英才</a>的校招平台最受重视，因为很多公司的校招平台都用的是中华英才的，但是各种求职网站校招简历和社招简历需要分别提交也是很明显的问题；<br />
分享： 本校的BBS论坛， 北京地区IT类里面：北邮人，水木都非常火爆，各种BBS也是其他面试的学生面试前参考信息源；</p>]]></description>
         <link>http://www.chedong.com/blog/archives/001484.html</link>
         <guid>http://www.chedong.com/blog/archives/001484.html</guid>
                  <category domain="http://www.sixapart.com/ns/types#category">工作招聘 |recruit</category>
        
        
         <pubDate>Fri, 29 Oct 2010 20:16:22 +0800</pubDate>
      </item>
            <item>
         <title>MT上“Name &quot;Locale::Maketext::Lexicon&quot; used only once:” 问题的解决： 改用Perl内置函数库</title>
         <description><![CDATA[<p>最近从服务器日志中经常发现<a href="http://www.chedong.com/blog/archives/000001.html">MT</a>的错误日志：<br />
Name "Locale::Maketext::Lexicon" used only once: possible typo at<br />
 .../extlib/Locale/Maketext.pm line 653.,....</p>

<p>解决方法：<br />
删除 mt/extlib/目录下的I18N/ 和 Locale/ 目录即可；</p>

<p>原因： <br />
服务器上已经升级到Perl 5.10.x了： 很多国际化字符集方面的支持直接使用Perl内置的支持包即可。<br />
[chedong@titans ~/logs/chedong.com/http]$ perl -v<br />
<blockquote>This is perl, v5.10.0 built for x86_64-linux-gnu-thread-multi</p>

<p>Copyright 1987-2007, Larry Wall</p>

<p>Perl may be copied only under the terms of either the Artistic License or the<br />
GNU General Public License, which may be found in the Perl 5 source kit.</p>

<p>Complete documentation for Perl, including FAQ lists, should be found on<br />
this system using "man perl" or "perldoc perl".  If you have access to the<br />
Internet, point your browser at http://www.perl.org/, the Perl Home Page.</blockquote></p>]]></description>
         <link>http://www.chedong.com/blog/archives/001483.html</link>
         <guid>http://www.chedong.com/blog/archives/001483.html</guid>
                  <category domain="http://www.sixapart.com/ns/types#category">MovableType|MT</category>
        
        
         <pubDate>Sun, 01 Aug 2010 17:44:16 +0800</pubDate>
      </item>
            <item>
         <title> 如何确定抽样统计的最小样本量（附：随机抽样统计的抽样误差Excel计算表格）</title>
         <description><![CDATA[<p>在电视节目中经常看到关于选举的报道中经常会后有支持率的数字，例如：调查结果为</p><ul><li>a方支持率为45.3%；</li><li>b方支持率为30.2%； </li><li>c方支持率为8.5%；</li><li>...<br /></li></ul><p>最后都会说明一下，此次电话调查的数量2352，置信度为95%﹐最大抽样误差为&plusmn;2.5%。</p><p>抽样调查的典型情景：对一个大的集合（比如：数千万选民）做一次调查的成本较高，抽样调查可以低成本的用近似的（可接受的）数据反映实际情况；在用户调研中，也经常通过通过抽样调查的方式并对比打分的方法做评估。这里就需要了解置信度和抽样误差的概念；  </p><p>抽样误差： 假如相同规模的抽样调查进行多次， 抽样均值在真实均值的上下波动，相对于整体均值的偏移波动就是抽样误差，而这个误差的分布是符合<a target="_blank" href="http://zh.wikipedia.org/zh-cn/%E6%A0%87%E5%87%86%E6%AD%A3%E6%80%81%E5%88%86%E5%B8%83">标准正态分布</a>的，例如下图： 横轴为整体的均值，圆点是每次抽样的均值，而红色那次抽样就是加上误差后都未覆盖到均值线的情况）；&nbsp;  </p><p><a title="原图作者： 崔玮" href="http://t.sina.com.cn/1250608490"><img border="0" src="http://www.chedong.com/blog/archives/2010/07/17/normal_distribution.png" /></a>&nbsp;</p><p>最小抽样量的计算公式： 抽样量需要 &gt; 30个才算足够多，可以用以下近似的误差/样本量估算公式； <br /> <a target="_blank" href="http://www.codecogs.com/eqnedit.php?latex=%5Cfn_jvn%20%5C120dpi%20n=%5Cfrac%7B%28%7Bz_%7B%5Calpha%20/2%7D%7D%29%5E%7B2%7D%7B%5Csigma%7D%5E2%7D%7BE%5E2%7D"><img border="0" src="http://latex.codecogs.com/png.latex?%5Cfn_jvn%20%5C120dpi%20n%5Capprox%5Cfrac%7B%28%7Bz_%7B%5Calpha%20/2%7D%7D%29%5E%7B2%7D%7B%5Csigma%7D%5E2%7D%7BE%5E2%7D" /></a> <br />n： 为样本量；<br /> <a target="_blank" href="http://www.codecogs.com/eqnedit.php?latex=%5Cfn_jvn%20%5C120dpi%20%7B%5Csigma%7D%5E2"><img width="18" height="16" border="0" alt="\fn_jvn \120dpi {\sigma}^2" title="\fn_jvn \120dpi {\sigma}^2" src="http://latex.codecogs.com/png.latex?%5Cfn_jvn%20%5C120dpi%20%7B%5Csigma%7D%5E2" /></a>：<a target="_blank" href="http://zh.wikipedia.org/zh-cn/%E6%96%B9%E5%B7%AE">方差</a>，抽样个体值和整体均值之间的偏离程度，抽样数值分布越分散方差越大，需要的采样量越多；<br />E： 为抽样误差（可以根据均值的百分比设定），由于是倒数平方关系，抽样误差减小为1/2，抽样量需要增加为4倍； <br /><a target="_blank" href="http://www.codecogs.com/eqnedit.php?latex=%5Cfn_jvn%20%5C120dpi%20%5E%7Bz_%7B%5Calpha/2%7D%7D"><img border="0" alt="\fn_jvn \120dpi ^{z_{\alpha/2}}" title="\fn_jvn \120dpi ^{z_{\alpha/2}}" src="http://latex.codecogs.com/png.latex?%5Cfn_jvn%20%5C120dpi%20%5E%7Bz_%7B%5Calpha/2%7D%7D" /></a>: 为可靠性系数，即置信度，置信度为95%时，<a target="_blank" href="http://www.codecogs.com/eqnedit.php?latex=%5Cfn_jvn%20%5C120dpi%20%5E%7Bz_%7B%5Calpha/2%7D%7D"><img border="0" alt="\fn_jvn \120dpi ^{z_{\alpha/2}}" title="\fn_jvn \120dpi ^{z_{\alpha/2}}" src="http://latex.codecogs.com/png.latex?%5Cfn_jvn%20%5C120dpi%20%5E%7Bz_%7B%5Calpha/2%7D%7D" /></a>=1.96，置信度为90%时，<a target="_blank" href="http://www.codecogs.com/eqnedit.php?latex=%5Cfn_jvn%20%5C120dpi%20%5E%7Bz_%7B%5Calpha/2%7D%7D"><img border="0" alt="\fn_jvn \120dpi ^{z_{\alpha/2}}" title="\fn_jvn \120dpi ^{z_{\alpha/2}}" src="http://latex.codecogs.com/png.latex?%5Cfn_jvn%20%5C120dpi%20%5E%7Bz_%7B%5Calpha/2%7D%7D" /></a>=1.645，置信度越高需要的样本量越多；95%置信度比90%置信度需要的采样量多40%；<br /><br />为了体现相对差距： 假设抽样<a target="_blank" href="http://zh.wikipedia.org/zh-cn/%E7%AE%97%E6%9C%AF%E5%B9%B3%E5%9D%87%E6%95%B0">均值</a>为 y </p><p>相对抽样误差 h = E / y </p><p><a href="http://zh.wikipedia.org/zh-cn/%E5%8F%98%E5%BC%82%E7%B3%BB%E6%95%B0">变异系数</a> C= &sigma; / y<br /></p><p><a target="_blank" href="http://www.codecogs.com/eqnedit.php?latex=%5Cfn_jvn%20%5C120dpi%20n=%5Cfrac%7B%28%7Bz_%7B%5Calpha%20/2%7D%7D%29%5E%7B2%7D%7B%5Csigma%7D%5E2%7D%7BE%5E2%7D"><img border="0" src="http://latex.codecogs.com/png.latex?%5Cfn_jvn%20%5C120dpi%20n%5Capprox%5Cfrac%7B%28%7Bz_%7B%5Calpha%20/2%7D%7D%29%5E%7B2%7DC%5E2%7D%7Bh%5E2%7D" /></a> <br /></p><p>以下是基于抽样得分的抽样误差估算表格： 方差越大需要的样本量越多，数据离散度越低，需要的抽样量越少；<br /></p><p><table width="384pt" border="0" style="border-collapse: collapse; height: 76px">       <tbody><tr style="height: 19px"> <td align="center" style="color: #000000; font-size: 9pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: center; vertical-align: middle; height: 38px; width: 80px" rowspan="2" class="et6">置信度</td><td align="center" style="color: #000000; font-size: 9pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: center; vertical-align: middle; height: 19px; width: 464px" colspan="5" class="et6">相对抽样误差(假设：C=0.4)</td> </tr> <tr style="height: 19px"> <td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 112px" class="et9">1%</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et9">2%</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et9">3%</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et8">4%</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et8">5%</td> </tr> <tr style="height: 19px"> <td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: center; vertical-align: top; white-space: normal; border-left: 0.5pt solid #000000; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et5">95%</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 112px" class="et4">6147&nbsp;</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et4">1537&nbsp;</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et4">683&nbsp;</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et4">384&nbsp;</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et4">246&nbsp;</td> </tr> <tr style="height: 19px"> <td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: center; vertical-align: top; white-space: normal; border-left: 0.5pt solid #000000; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et5">90%</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 112px" class="et4">4330&nbsp;</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et4">1082&nbsp;</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et4">481&nbsp;</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et4">271&nbsp;</td><td align="center" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px" class="et4">173&nbsp;</td> </tr>  </tbody></table> </p><p>如果是基于胜出率，支持率等： 分值为0/1状态分布，公式拟合为<span style="font-size: 10.5pt; font-family: '宋体'" /></p><p style="margin-bottom: 0pt; margin-top: 0pt" class="p0"><img border="0" src="http://latex.codecogs.com/png.latex?%5Cfn_jvn%20n%5Capprox%5Cfrac%7B%28%7Bz_%7B%5Calpha%20/2%7D%7D%29%5E%7B2%7D%5Cpi%281-%5Cpi%29%7D%7BE%5E2%7D" />&nbsp; </p><p style="margin-bottom: 0pt; margin-top: 0pt" class="p0">&pi;为按照经验得出的最后比例，在未知时&pi;可取50%，待算出结果后再重新拟合，比例越悬殊需要的样本量越少；&nbsp; <br /></p><p style="margin-bottom: 0pt; margin-top: 0pt" class="p0">&nbsp; <!--td {color:#000000; font-size:12.0pt; font-family:宋体; font-weight:400; font-style:normal; text-decoration:none; text-align:general; vertical-align:middle; mso-number-format:General; mso-protection:locked visible; }.et9 {color:#000000; font-size:10.5pt; font-family:宋体; font-weight:700; font-style:normal; text-decoration:none; text-align:center; vertical-align:top; white-space:normal; border-left:.5pt solid #000000; border-right: .5pt solid #000000; border-bottom:.5pt solid #000000; mso-number-format:'0%'; mso-protection:locked visible; }.et10 {color:#000000; font-size:10.5pt; font-family:宋体; font-weight:700; font-style:normal; text-decoration:none; text-align:center; vertical-align:top; border-right: .5pt solid #000000; border-top:.5pt solid #000000; border-bottom:.5pt solid #000000; mso-number-format:General; mso-protection:locked visible; }.et11 {color:#000000; font-size:10.5pt; font-family:宋体; font-weight:400; font-style:normal; text-decoration:none; text-align:justify; vertical-align:top; white-space:normal; border-right: .5pt solid #000000; border-bottom:.5pt solid #000000; mso-number-format:General; mso-protection:locked visible; }.et12 {color:#000000; font-size:10.5pt; font-family:宋体; font-weight:700; font-style:normal; text-decoration:none; text-align:general; vertical-align:top; white-space:normal; border-right: .5pt solid #000000; border-bottom:.5pt solid #000000; mso-number-format:'0%'; mso-protection:locked visible; }.et13 {color:#000000; font-size:10.5pt; font-family:宋体; font-weight:700; font-style:normal; text-decoration:none; text-align:general; vertical-align:top; white-space:normal; border-right: .5pt solid #000000; border-top:.5pt solid #000000; border-bottom:.5pt solid #000000; mso-number-format:'0%'; mso-protection:locked visible; }.et14 {color:#000000; font-size:10.5pt; font-family:宋体; font-weight:700; font-style:normal; text-decoration:none; text-align:center; vertical-align:top; border-right: .5pt solid #000000; border-top:.5pt solid #000000; border-bottom:.5pt solid #000000; mso-number-format:General; mso-protection:locked visible; }.et15 {color:#000000; font-size:9.0pt; font-family:宋体; font-weight:700; font-style:normal; text-decoration:none; text-align:center; vertical-align:middle; border-left:.5pt solid #000000; border-right: .5pt solid #000000; border-top:.5pt solid #000000; mso-number-format:General; mso-protection:locked visible; }.et16 {color:#000000; font-size:9.0pt; font-family:宋体; font-weight:700; font-style:normal; text-decoration:none; text-align:center; vertical-align:middle; border-left:.5pt solid #000000; border-right: .5pt solid #000000; border-bottom:.5pt solid #000000; mso-number-format:General; mso-protection:locked visible; }-->  </p><table width="360pt" border="0" style="border-collapse: collapse; height: 76px">       <tbody><tr style="height: 19px">  <td class="et15" rowspan="2" style="color: #000000; font-size: 9pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: center; vertical-align: middle; border: 0.5pt solid #000000; height: 38px; width: 80px">置信度</td><td class="et10" colspan="5" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: center; vertical-align: top; border-right: 0.5pt solid #000000; border-top: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 400px">相对抽样误差</td>  </tr> <tr style="height: 19px"> <td align="center" class="et12" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">1%</td><td align="center" class="et13" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-top: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">2%</td><td align="center" class="et13" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-top: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">3%</td><td align="center" class="et13" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-top: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">4%</td><td align="center" class="et13" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-top: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">5%</td> </tr> <tr style="height: 19px"> <td class="et9" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: center; vertical-align: top; white-space: normal; border-left: 0.5pt solid #000000; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">95%</td><td align="right" class="et11" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">9604</td><td align="right" class="et11" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">2401</td><td align="right" class="et11" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">1067</td><td align="right" class="et11" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">600</td><td align="right" class="et11" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">384</td> </tr> <tr style="height: 19px"> <td class="et9" style="color: #000000; font-size: 10.5pt; font-weight: 700; font-style: normal; text-decoration: none; text-align: center; vertical-align: top; white-space: normal; border-left: 0.5pt solid #000000; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">90%</td><td align="right" class="et11" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">6765</td><td align="right" class="et11" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">1691</td><td align="right" class="et11" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">752</td><td align="right" class="et11" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">423</td><td align="right" class="et11" style="color: #000000; font-size: 10.5pt; font-weight: 400; font-style: normal; text-decoration: none; text-align: justify; vertical-align: top; white-space: normal; border-right: 0.5pt solid #000000; border-bottom: 0.5pt solid #000000; height: 19px; width: 80px">270</td> </tr>  </tbody></table> <p>从而看出大部分的电话抽样调查：95%置信度的情况下，误差要控制在2%以内取样量一般在2000-5000；为了方便计算抽样调查的误差和估算抽样量，制作了一个Excel表格附后，调整颜色框中的抽样量数字就可以得到相应的误差或根据指定的误差范围估算出抽样量；<br /></p>]]></description>
         <link>http://www.chedong.com/blog/archives/001462.html</link>
         <guid>http://www.chedong.com/blog/archives/001462.html</guid>
                  <category domain="http://www.sixapart.com/ns/types#category">日志分析|Analytics</category>
        
        
         <pubDate>Sat, 17 Jul 2010 23:04:14 +0800</pubDate>
      </item>
            <item>
         <title>AWStats 7.0 发布， Linux下的安装配置使用备忘</title>
         <description><![CDATA[<p>5月25日，<a href="http://www.amazon.com/exec/obidos/wishlist/3F19UHYYQRBPA">Eldy</a>发布了AWStats的7.0版本，从界面上看主要的变化是很多统计图使用了<a href="http://code.google.com/intl/zh-CN/apis/charttools/">Google图表API</a>生成；<br />
<img alt="awstats_70_googlechartapi.png" src="http://www.chedong.com/blog/archives/2010/06/05/awstats_70_googlechartapi.png" width="681" height="552" /></p>

<p>在Linux的安装请参考<a href="http://www.chedong.com/tech/awstats.html">AWStats安装笔记</a>， 此次更新后的相关配置修改都放到<a href="http://www.chedong.com/tech/lib.tgz">补充库</a>里了；<br />
包含chedong.com的awstats配置样例：awstats.chedong.conf <br />
通用配置文件样例： coommon.conf<br />
搜索引擎和蜘蛛定义更新： robots.pm search_engines.pm （需要放到lib目录下）</p>

<p>缺省建议启用的配置：common.conf<br />
LoadPlugin="tooltips"<br />
LoadPlugin="decodeutfkeys"<br />
<strong>LoadPlugin="graphgooglechartapi"</strong><br />
LoadPlugin="<a href="http://www.chedong.com/blog/archives/001182.html">geoip</a> GEOIP_STANDARD /home/chedong/chedong.com/cgi-bin/awstats/GeoIP.dat"<br />
LoadPlugin="geoip_city_maxmind GEOIP_STANDARD /home/chedong/chedong.com/cgi-bin/awstats/GeoLiteCity.dat"<br />
LoadPlugin="<a href="http://www.chedong.com/blog/archives/001356.html">qqhostinfo</a>"</p>

<p>搜索引擎定义主要去掉了一些门户网站的域名（例如： 163.com live.com），避免了搜索服务和门户其他服务(例如： 博客)容易和搜索来源混淆的情况。<br />
蜘蛛定义主要增加了feed抓取机器人： <br />
'sosospider','<a href="http://help.soso.com/webspider.htm"  target="_blank">SosoSpider</a>',<br />
'youdaobot','<a href="http://www.youdao.com/help/webmaster/spider/"  target="_blank">YoudaoBot</a>',<br />
'doubanbot','<a href="http://www.douban.com"  target="_blank">DoubanBot</a>',<br />
'friendfeedbot','<a href="FriendFeedBot"  target="_blank">FriendFeedBot</a>',</p>

<p>AWStats 7.0的详细ChangeLog更新附后：</p>]]></description>
         <link>http://www.chedong.com/blog/archives/001480.html</link>
         <guid>http://www.chedong.com/blog/archives/001480.html</guid>
                  <category domain="http://www.sixapart.com/ns/types#category">日志分析|AWStats</category>
        
        
         <pubDate>Sat, 05 Jun 2010 15:38:35 +0800</pubDate>
      </item>
      
   </channel>
</rss>

