Supplement 1. Raw survey results and data cleaning/processing scripts for R.

File List DataCleaning_20120105.R (md5:7c3b115561b5c687e34c7ffa63896a82)<br><br> DataTrimming_20120105.R (md5:77f0bbe669da955c81d9b5fb2d3e2d3e)<br><br> AdvancedSpreadsheet_20110526.csv (md5:db607c6c3238ffb45effa2c97ec40b9a)<br><br> VariableRenaming_20110711.csv (md5:ac7519d78d3eac3d2ccf82ef94f1af80)<br><br> ParamTable.csv (md5:ff3fa18d83a3a6b860462c2955a5c8c9) Description <i>How the files should be used:</i> Scripts (2) DataCleaning_20120105.R is an R script for cleaning up data prior to analysis. This script removes spaces, substitutes text for codes, removed duplicate schools, and converts questions and answers from the survey into more simple parameter names, without any numbers, spaces, or symbols. This script is heavily annotated to assist the user of the file in understanding what is being done to the data files. The script produces the file cleandata_[date].Rdata, which is called in the file DataTrimming_20120105.R DataTrimming_20120105.R is an R script for trimming extraneous variables not used in final analyses. Some variables are combined as needed and NAs (no answers) are removed. The file is heavily annotated. It produces trimdata_[date].Rdata, which was imported into Excel for summary statistics. Data files (3) AdvancedSpreadsheet_20110526.csv is the output file from the SurveyMonkey online survey tool used for this project. It is a .csv sheet with the complete set of survey data, although some data (e.g., open-ended responses, institution names) are removed to prevent schools and/or instructors from being identifiable. This file is read into DataCleaning_20120105.R for cleaning and editing. VariableRenaming_20110711.csv is called into the DataCleaning_20120105.R script to convert the questions and answers from the survey into simple parameter names, without any numbers, spaces, or symbols. ParamTable.csv is a list of the parameter names used for analysis and the value codes. It can be used to understand outputs from the scripts above (cleandata_[date].Rdata and trimdata_[date].Rdata).

文件清单：<br><br>DataCleaning_20120105.R（md5:7c3b115561b5c687e34c7ffa63896a82）<br><br>DataTrimming_20120105.R（md5:77f0bbe669da955c81d9b5fb2d3e2d3e）<br><br>AdvancedSpreadsheet_20110526.csv（md5:db607c6c3238ffb45effa2c97ec40b9a）<br><br>VariableRenaming_20110711.csv（md5:ac7519d78d3eac3d2ccf82ef94f1af80）<br><br>ParamTable.csv（md5:ff3fa18d83a3a6b860462c2955a5c8c9）<br><br>文件使用说明：<br><br>脚本文件（共2个）：<br>1. DataCleaning_20120105.R为用于分析前数据清洗的R脚本。该脚本可移除数据中的空格、将文本替换为编码、删除重复院校，并将调查问卷的问题与答案转换为不含数字、空格或符号的简洁参数名。本脚本附带详细注释，便于使用者理解其对数据文件的处理流程。脚本执行后将生成cleandata_[日期].Rdata文件，该文件会被DataTrimming_20120105.R调用。<br>2. DataTrimming_20120105.R为用于剔除最终分析中未使用的冗余变量的R脚本，必要时会合并部分变量并移除缺失值（NAs，即无作答项）。该脚本同样附带详细注释，执行后将生成trimdata_[日期].Rdata文件，该文件可导入Excel以计算描述性统计量。<br><br>数据文件（共3个）：<br>1. AdvancedSpreadsheet_20110526.csv：本项目所用在线调查工具（SurveyMonkey）生成的输出文件，为包含完整调查数据集的.csv格式文件。为避免泄露院校及授课教师身份，部分数据（如开放式回复内容、院校名称）已被移除。该文件会被读入DataCleaning_20120105.R中进行清洗与编辑。<br>2. VariableRenaming_20110711.csv：会被DataCleaning_20120105.R脚本调用，用于将调查问卷的问题与答案转换为不含数字、空格或符号的简洁参数名。<br>3. ParamTable.csv：收录了本次分析所用的参数名及其对应值编码，可用于理解上述脚本生成的cleandata_[日期].Rdata与trimdata_[日期].Rdata文件的输出结果。