ISSS608 2016-17 T1 Assign2 ZHANG Zhe

Introduction

Wikipedia and education often go together. Wikis are sometimes used in classrooms for collaborative projects. Some teachers have found, however, that learners prefer to add their own content rather than rewrite others' work, perhaps because of an institutionally cultivated norm of individual ownership. Some students also express shyness about exposing their work to be viewed by others. Such transparency seems to reduce plagiarism. In addition, Wikis also could help teacher to do lesson preparation. This research was held in 2015 and focus on university faculty perceptions and practices of using Wikipedia as a teaching resource. The survey items include 13 sections, Perceived Usefulness, Perceived Ease of Use, Perceived Enjoyment, Quality, Visibility, Social Image, Sharing Attitude, Use Behavior, Profile 2.0, Job Relevance, Behavioral Intention, Incentives and Experience.

Data Preparation

Data Format

• Missing Data

In the original file, missing data was expressed by symbol ‘?’. However, other data are all numbers, and it would be difficult to deal with the whole file later. Wiki4HE.csv was opened in JMP with preview, then check ‘Other’ in ‘End of Field’ and input English character ‘;’. The next step is replacing all missing data ‘?’ by space. Finally save it.

• Data Type

Although some demographic information shows numbers, these numbers represent specific meanings. When the file was imported into Tableau, all variables’ data types were defaulted as number(whole). According to the attribute information, types of GENDER, DOMAIN, PhD, UNIVERSITY, UOC_POSITION, OTHER, OTHER_POSITION and USERWIKI were changed to string.

• Puzzling Information

In the file imported, Other Position has 2 levels, ‘1’ and ‘2’. Otherstatus has 7 levels, from 1 to 7. These information conflicts with data set attribute description on the website. To fit the dataset description, we modify the 2 columns names on the Tableau to Other and Other Position, respectively. Data Content

• Data Integration

There are 13 sections in this survey. According to details of each section, we could do the further classification to analyze better and visual more friendly. The specific process is shown as below. There are 4 parts now. Pivot all survey questions and change the 2 columns to Survey Items and Score, separately. Right click Survey Items and choose Create → Group, then group relative questions based on the dataset description. Finally change column name to Sections. Repeat the previous step on 13 Section by the classification as stated above. And name it Parts.