Difference between revisions of "IS428 AY2019-20T2 Assign LU ZHIMAO"
(64 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
+ | [[File:C8tpsT8UMAAN3c .jpg|700px|thumb|right]] | ||
+ | =<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Project Motivation'''</div>= | ||
+ | |||
+ | Every two years, SMU Libraries conducts a comprehensive survey in which faculty, students and staff have the opportunity to rate various aspects of our services. The survey provides SMU Libraries with input to help enhance existing services and to anticipate emerging needs of SMU faculty, students and staff. | ||
+ | |||
+ | However, based on the current report provided by the library it requires readers to complete 155 pages of report in order to have a full understanding of the content. After reading through all the pages of the report, the report only provides a surface level of analysis in which the library cannot draw much insightful information out of it. | ||
+ | |||
+ | Thus, in order to have better visualization and understanding of the data from the survey results, I have decided to use Tableau as a platform which allows the Library to toggle between the various groups; Undergraduate, Postgraduate, Staff and Faculty and also offer several types of data visualization that can be more easily readable. | ||
+ | |||
+ | =<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Project Objectives and task'''</div>= | ||
+ | |||
+ | In this assignment, the aim is to apply 10 Weeks of Visual Analytics for Business Intelligence module's knowledge with self-learning techniques to help Singapore Management University Library to deliver a focused and compact visualisation. By doing, it allows Singapore Management University Library to be well-informed of their service feedback from users as well as to identify those critical areas currently in the needs of improvement in order to satisfy user's experience. | ||
− | + | Using visual analytics approach to reveal the level of services provided by SMU libraries as perceived by: | |
− | + | * the undergraduate students, | |
+ | * the postgraduate students, | ||
+ | * the faculty, | ||
+ | * the staff. | ||
=<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Data Preparation'''</div>= | =<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Data Preparation'''</div>= | ||
− | + | =='''Data Cleaning Using Excel and Tableau'''== | |
+ | |||
+ | # Analysing the raw excel data, it becomes evident that there are multiple rows with data that are either an empty space " " or "NIL". Since both of them are referring to the same output. In order not to confuse the readers, I have decided to combine both data together and label it as "no input". Hence in ranking columns, it would show 1 to 7 and "no input". | ||
+ | # Under the question "Which library do you use most", the answer is reflected on the excel sheet as "1" or "2" where "1" is a representation of "Li Ka Shing Library" and "2" is a representation of "Kwa Geok Choo Library". This can be quite confusing, as the usage of numbers is to represent the ranking but it is only exceptional in this case. Therefore, I have decided to convert those relevant data into a word format instead of it being represented by numerical numbers. | ||
+ | {| class="wikitable" | ||
+ | |- | ||
+ | ! <span style="color:red">Before</span> | ||
+ | |- | ||
+ | | | ||
+ | [[File:Qqqq.jpg|1000px|thumb|center]] | ||
+ | [[File:Dfjbdu.jpg|900px|thumb|center]] | ||
+ | |- | ||
+ | ! <span style="color:red">After</span> | ||
+ | [[File:111qqq.jpg|1000px|thumb|center]] | ||
+ | |- | ||
+ | |} | ||
+ | |||
+ | # Tableau prefers data to be formatted in a machine-readable format. To reshape data for easier analysis in Tableau, I need to perform a pivot. In Tableau, pivoting means transposing data from a crosstab format into a columnar format–from wide, short tables into thin, tall tables. Tableau Prep Builder makes pivoting visual, allowing to see how data changes with every step. | ||
+ | ## Connect to the library data source | ||
+ | ## Drag the table that you want to pivot | ||
+ | ## Click the down arrow icon and select Pivot from the context menu | ||
+ | ## After selecting, the results appear immediately in the pivoted form | ||
+ | {| class="wikitable" | ||
+ | |- | ||
+ | ! <span style="color:red">Before</span> | ||
+ | |- | ||
+ | | | ||
+ | [[File:Bhh.jpg|1000px|thumb|center]] | ||
+ | |- | ||
+ | ! <span style="color:red">After</span> | ||
+ | [[File:B.jpg|1000px|thumb|center]] | ||
+ | |- | ||
+ | |} | ||
− | + | # After looking through the pivoted data and comparing with the report provided, I realised that the category types are missing in this data. These categories are important as it allows those relevant questions to group under the same category. So that it allows the users to choose how deep they prefer their data to be in terms of presenting the data visualization which allows the use not only limits to question along. With that, I will be extracting the latest pivoted data set from Tableau and saved as a csv file. On the csv file, I will create a new column and label it "Category". Under that column, I will insert the relevant grouping titles; Communication, Facilities and Equipment, Information Resources and Service Delivery. This is in accordance with the groupings as by the report provided. | |
+ | |||
+ | {| class="wikitable" | ||
+ | |- | ||
+ | ! <span style="color:red">Before</span> | ||
+ | |- | ||
+ | | | ||
+ | [[File:888.jpg|1000px|thumb|center]] | ||
+ | [[File:9999.jpg|1000px|thumb|center]] | ||
+ | |- | ||
+ | ! <span style="color:red">After</span> | ||
+ | |- | ||
+ | | | ||
+ | [[File:99999999.jpg|900px|thumb|center]] | ||
+ | |} | ||
=='''Selection of Graphs used in Data Visualization'''== | =='''Selection of Graphs used in Data Visualization'''== | ||
Line 16: | Line 77: | ||
! Graph Type | ! Graph Type | ||
! Reasoning | ! Reasoning | ||
− | ! | + | ! Type of Graph would like to use in library survey data analysis |
|- | |- | ||
| <span style="color:red">Likert Scale</span> | | <span style="color:red">Likert Scale</span> | ||
− | | They are one of the most reliable ways to measure opinions, perceptions and Behaviours. It is used to an analysis survey question that uses a 5 or 7-point scale, in our Survey Data it uses 7-point scale and it referred to as a satisfaction scale, that ranges from Low (represented by 1) to High 7 (represented by 7). | + | | |
+ | * They are one of the most reliable ways to measure opinions, perceptions and Behaviours. It is used to an analysis survey question that uses a 5 or 7-point scale, in our Survey Data it uses 7-point scale and it referred to as a satisfaction scale, that ranges from Low (represented by 1) to High 7 (represented by 7). | ||
| | | | ||
− | [[File: Likert scale.png| | + | [[File: Likert scale.png|600px|thumb|Example of Likert Scale Chart |
Source:https://www.researchgate.net/profile/Ron_Fricker/publication/305215522/figure/fig13/AS:383031459237890@1468333224542/An-example-of-a-diverging-stacked-bar-chart-for-a-five-point-Likert-scale-question-This.png]] | Source:https://www.researchgate.net/profile/Ron_Fricker/publication/305215522/figure/fig13/AS:383031459237890@1468333224542/An-example-of-a-diverging-stacked-bar-chart-for-a-five-point-Likert-scale-question-This.png]] | ||
|- | |- | ||
| <span style="color:red">KPI Performance Chart</span> | | <span style="color:red">KPI Performance Chart</span> | ||
− | | The KPI chart is used to, at a quick glance, give information about the current performance of the library. KPIs can be used to track the performance of the library against importance. They can provide a management tool for gaining insight and decision making. For our library survey data analysis is to identify the gaps between the importance and performance and in this case, both indicators are set by the survey responders. By looking at the gap we are able to identify if the library services currently under user's expectation or achieved user's expectation. | + | | |
+ | * The KPI chart is used to, at a quick glance, give information about the current performance of the library. KPIs can be used to track the performance of the library against importance. They can provide a management tool for gaining insight and decision making. For our library survey data analysis is to identify the gaps between the importance and performance and in this case, both indicators are set by the survey responders. By looking at the gap we are able to identify if the library services currently under user's expectation or achieved user's expectation. | ||
| | | | ||
− | [[File:KIP.png| | + | [[File:KIP.png|600px|thumb|Example of KPI Performance Chart |
Source:https://lh3.googleusercontent.com/-c8CKjPabp1g/V3FJWKo3KLI/AAAAAAAAyog/B8nHUvLW8tA/Screen-Shot-2016-06-27-at-4.31.26-pm.png?imgmax=9999]] | Source:https://lh3.googleusercontent.com/-c8CKjPabp1g/V3FJWKo3KLI/AAAAAAAAyog/B8nHUvLW8tA/Screen-Shot-2016-06-27-at-4.31.26-pm.png?imgmax=9999]] | ||
|- | |- | ||
| <span style="color:red">Vertical Bar Chart</span> | | <span style="color:red">Vertical Bar Chart</span> | ||
− | | Use vertical column charts to display ordinal variables. For Net Promoter Score, it arranges ordinal categories from left to right so readers can view the sequence accordingly. | + | | |
+ | * Use vertical column charts to display ordinal variables. For Net Promoter Score, it arranges ordinal categories from left to right so readers can view the sequence accordingly. | ||
| | | | ||
− | [[File:Netpromoter-summary.png.png| | + | [[File:Netpromoter-summary.png.png|600px||thumb|Example of Vertical Bar Chart |
Source:https://images.ctfassets.net/freurdme1ae3/xQwhSQHldq9sM3FuEuXLC/04a4b44eb7c87b6589e99c54acdd6e31/uploadssites1net-promoter-score-summary.jpg?fm=jpg&w=1480&q=70]] | Source:https://images.ctfassets.net/freurdme1ae3/xQwhSQHldq9sM3FuEuXLC/04a4b44eb7c87b6589e99c54acdd6e31/uploadssites1net-promoter-score-summary.jpg?fm=jpg&w=1480&q=70]] | ||
|- | |- | ||
| <span style="color:red">Horizontal Bar Chart</span> | | <span style="color:red">Horizontal Bar Chart</span> | ||
− | | Use horizontal bar charts to display nominal variables like Question ID. It allows arranging Question ID in a list from top to bottom and in a horizontal manner. | + | | |
+ | * Use horizontal bar charts to display nominal variables like Question ID. It allows arranging Question ID in a list from top to bottom and in a horizontal manner. | ||
| | | | ||
[[File:TextLabel2007Bar08.png|thumb|center|900px||Example of Horizontal Bar Chart | [[File:TextLabel2007Bar08.png|thumb|center|900px||Example of Horizontal Bar Chart | ||
Line 43: | Line 108: | ||
|- | |- | ||
| <span style="color:red">Population Pyramid Chart</span> | | <span style="color:red">Population Pyramid Chart</span> | ||
− | |For Population Pyramid chart, breaks down SMU population into positions such as Undergraduate, Postgraduate, Staff and Faculty. For this assignment, you'll find the left side of the pyramid graphing the LKS population and the right side of the pyramid displaying the KGC population. | + | | |
+ | * For Population Pyramid chart, breaks down SMU population into positions such as Undergraduate, Postgraduate, Staff and Faculty. For this assignment, you'll find the left side of the pyramid graphing the LKS population and the right side of the pyramid displaying the KGC population. | ||
| | | | ||
− | [[File:Pyramid chart.png| | + | [[File:Pyramid chart.png|600px|thumb|Example of Pyramid Chart |
Source:https://www.infragistics.com/community/resized-image/__size/1040x924/__key/communityserver-blogs-components-weblogfiles/00-00-00-09-43/2018_2D00_05_2D00_16_5F00_12_2D00_09_2D00_33.png]] | Source:https://www.infragistics.com/community/resized-image/__size/1040x924/__key/communityserver-blogs-components-weblogfiles/00-00-00-09-43/2018_2D00_05_2D00_16_5F00_12_2D00_09_2D00_33.png]] | ||
+ | |- | ||
+ | | <span style="color:red">Highlight Table</span> | ||
+ | | | ||
+ | *The highlight table allows us to apply conditional formatting to a view. Tableau will automatically apply a colour scheme in either a continuous or stepped array of colours from highest to lowest. It is great for comparing a field’s values within a row or column. | ||
+ | | | ||
+ | [[File:Highlight-table-demonstrating-the-percentage-of-total-student-numbers-visiting-Chester.png|600px|thumb|center|Example of Highlight Table | ||
+ | Source:https://www.researchgate.net/profile/Andrew_Moss5/publication/293195352/figure/fig2/AS:329962755248141@1455680659656/Highlight-table-demonstrating-the-percentage-of-total-student-numbers-visiting-Chester.png]] | ||
+ | |- | ||
+ | | <span style="color:red">Treemap</span> | ||
+ | | | ||
+ | * A treemap is a visualization that nests rectangles in hierarchies so that can compare different dimension combinations across one or two measures (one for size; one for colour) and quickly interpret their respective contributions to the whole. | ||
+ | * By looking at the colour and the size of the rectangles represented by treemap, we able to identify the category and question quickly based on user's input. | ||
+ | | | ||
+ | [[File:Treemap.png|600px|thumb|center|Example of Treemap | ||
+ | Source:https://datavizcatalogue.com/methods/images/top_images/treemap.png]] | ||
|} | |} | ||
=<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Survey DashBoard Overview Information'''</div>= | =<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Survey DashBoard Overview Information'''</div>= | ||
+ | =='''Tableau Visualization'''== | ||
+ | The interactive Tableua visualization can access [https://public.tableau.com/profile/lu.zhimao#!/vizhome/Lu_Zhimao_Assignment/OverviewStory here]. | ||
=='''DashBoard Overview'''== | =='''DashBoard Overview'''== | ||
+ | [[File:Screenshot (30).png|900px|center]] | ||
+ | [[File:Screenshot (31).png|900px|center]] | ||
+ | [[File:Screenshot (32).png|900px|center]] | ||
+ | [[File:Screenshot (33).png|900px|center]] | ||
+ | |||
=='''DashBoard Summary'''== | =='''DashBoard Summary'''== | ||
{| class="wikitable" | {| class="wikitable" | ||
Line 76: | Line 164: | ||
|- | |- | ||
| 3 | | 3 | ||
− | | | + | | |
− | + | [[File:Screenshot (34).png|850px|thumb|center]] | |
− | + | * By using the Likert Scale we are able to group the answers into 3 categories (low, neutral, high). Neutral will be the benchmark to divide low and high apart. The size of the neutral split evenly and the splitting point will be the point 0 also known as the benchmark point. | |
− | + | * The Likert Scale is presented in the form of category followed by the question which is split into importance and performance). | |
+ | * Although there is a benchmark on each bar but the reader unable to have a clear comparison of each ranking between importance and performance. As the reader has to manually compare both figures provided. | ||
+ | * Therefore, at the later part of the individual analysis, I have added in an extra KIP Chart to help readers to have a clear understanding of the differences between Importance and Performance. | ||
+ | |||
|} | |} | ||
− | =<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Introduction'''</div>= | + | =<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Introduction for Gap Analysis'''</div>= |
=='''Gap Analysis Explanation'''== | =='''Gap Analysis Explanation'''== | ||
* The highlight table shows the percentage breakdown of rating for both importance and performance for each group – Undergraduate, Postgraduate, Staff and Faculty. | * The highlight table shows the percentage breakdown of rating for both importance and performance for each group – Undergraduate, Postgraduate, Staff and Faculty. | ||
− | * Conducting a gap analysis where we take the Performance minus the Importance to find the gap between the importance of the question to them and compare it to how the library performs. | + | * Since, the individual breakdown will be represented in the form of KIP Chart. For every KIP chart, there will be a KIP pre-set in order to conduct the comparison. In this case, the KIP will be the Importance figure set by the responders themselves which will be representing in the red line and the performance figure will be represented in the form of bar chart. |
+ | * Conducting a gap analysis where we take the Performance minus the Importance to find the gap between the importance of the question to them and compare it to how the library performs. | ||
* Since Importance is defined as how important it is to that individual and performance is defined as how well the library is performing in that area. | * Since Importance is defined as how important it is to that individual and performance is defined as how well the library is performing in that area. | ||
* We will take the difference between these two, to measure how well the library is performing when compared to the level of importance held to them. | * We will take the difference between these two, to measure how well the library is performing when compared to the level of importance held to them. | ||
* If the gap analysis turns out positive, it means that performance has exceeded importance. If it turns out negative, it means that performance has not met the same ranking as importance. And this is one key aspect the library should look at when improving their services. | * If the gap analysis turns out positive, it means that performance has exceeded importance. If it turns out negative, it means that performance has not met the same ranking as importance. And this is one key aspect the library should look at when improving their services. | ||
* I will be picking out 1 major change in each group for each category – Undergraduate, Postgraduate, Staff and Faculty and the rest can be viewed on Tableau Public. | * I will be picking out 1 major change in each group for each category – Undergraduate, Postgraduate, Staff and Faculty and the rest can be viewed on Tableau Public. | ||
− | |||
=<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Undergraduate Student'''</div>= | =<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Undergraduate Student'''</div>= | ||
Line 131: | Line 222: | ||
=='''Examining Correlation'''== | =='''Examining Correlation'''== | ||
− | [[File:Capturew.jpg| | + | [[File:Capturew.jpg|850px|thumb|center]] |
* Using a highlight table to examine the correlation between library visits and campus visits, we can see the percentage breakdown of the student’s frequency. | * Using a highlight table to examine the correlation between library visits and campus visits, we can see the percentage breakdown of the student’s frequency. | ||
* The table shows that although 25% indicated that they visit the school campus quarterly while also visiting the library daily. | * The table shows that although 25% indicated that they visit the school campus quarterly while also visiting the library daily. | ||
Line 138: | Line 229: | ||
=='''Library Frequency Visits'''== | =='''Library Frequency Visits'''== | ||
− | [[File:Screenshot (24).png| | + | |
+ | [[File:Screenshot (24).png|1000px|thumb|center]] | ||
* It can be seen from the breakdown of the library visits by undergraduates would be that majority of Business students patronize the library more than other schools. And the highest visits would be the weekly visits of business students across all schools at 42.44%. | * It can be seen from the breakdown of the library visits by undergraduates would be that majority of Business students patronize the library more than other schools. And the highest visits would be the weekly visits of business students across all schools at 42.44%. | ||
+ | |||
+ | =='''Trend of Missing Data'''== | ||
+ | [[File:Screenshot (27).png|1000px|thumb|center]] | ||
+ | * It is evident that under “Service Delivery”, the majority of Undergraduate students did not answer some of the questions in that section as compared to the rest. This is especially so for question 7 and question 9 where 12.09% and 9.63% did not answer the question. | ||
+ | * The questions that are generally left unanswered can be seen as irrelevant to the students as they probably never used that “service” or they do not require the library to provide them with such services. | ||
=<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Postgraduate Student'''</div>= | =<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Postgraduate Student'''</div>= | ||
Line 177: | Line 274: | ||
* The library should come out with an additional survey for these Postgraduates to feel the ground for the needs of these Postgraduates. | * The library should come out with an additional survey for these Postgraduates to feel the ground for the needs of these Postgraduates. | ||
|} | |} | ||
+ | |||
+ | =='''Library Frequency Visits'''== | ||
+ | [[File:Screenshot (25).png|1000px|thumb|center]] | ||
+ | * It can be seen that Postgraduates students that visit the library largely constitutes of business students. And the frequency of their visits is usually quarterly, at its highest of 64.71%. | ||
+ | * Whereas the percentage of visits, as well as the frequency of visits, is significantly lower for the other schools. | ||
+ | * The number of business students patronizing the library is almost double to triple the amount compared to other schools. | ||
+ | |||
+ | =='''Trend of Missing Data'''== | ||
+ | [[File:Screenshot (28).png|1000px|thumb|center]] | ||
+ | * Question 7 and Question 16 are seen to have the highest percentage of seeing no answers provided and again, this can be due to the nature of services or facilities which satisfies the needs of the target group. | ||
+ | * In the case of Postgraduates, it is evident that certain questions do not apply to them. | ||
+ | * For example, Question 16 is about “a computer is available when I need one”. This usually does not apply to Postgraduates as the majority would have had their own. | ||
=<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Faculty'''</div>= | =<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Faculty'''</div>= | ||
Line 212: | Line 321: | ||
* However, it is stated in the survey that the library is not performing up to par where there might be insufficient resources provided or lesser variety for them. | * However, it is stated in the survey that the library is not performing up to par where there might be insufficient resources provided or lesser variety for them. | ||
|} | |} | ||
+ | |||
+ | =='''Library Frequency Visits'''== | ||
+ | [[File:Screenshot (26).png|1000px|thumb|center]] | ||
+ | * From the data shown, we can see that the number of visits is evenly spread between 3 schools, namely; Accounting, Business and Economics. | ||
+ | * However, it can be seen that it is a consistent value of between 26%-29%. | ||
+ | * Law and Social Sciences do not even have the data for monthly, quarterly visits to the library, with the majority being daily and weekly. | ||
+ | |||
+ | =='''Trend of Missing Data'''== | ||
+ | [[File:Screenshot (29).png|1000px|thumb|center]] | ||
+ | * The high number of “no answers” is evidently flagged out under the “Facilities and Equipment ranging from 7-9%. | ||
+ | * With the second-highest coming from the category “Information Resources”, the highest at 5.34%. | ||
+ | * This shows that there is a large number of people under the Faculty category that did not answer these questions. | ||
+ | * As we can see those questions fall under “Facilities and Equipment” are generally asking about printing, photocopying facilities or can you find a quiet place for group project or study. As we know all the faculty member are able to get printing or photocopy facilities either from their own office or send for mass printing, which means that in most of the time they do not require such services. Not only that, but all the faculty members do also own an individual office to do their research or preparing for teaching material. Therefore, they do not require a place in the library. | ||
=<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Staff'''</div>= | =<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Staff'''</div>= | ||
Line 250: | Line 372: | ||
=<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Data Reference'''</div>= | =<div style="margin-top: 10px;font-family: Arial;font-weight:bold;text-align:center;font-size:18px; border: 5px solid #00000000; border-radius:8px;text-align:center; background-color: #ffef8e; color: black; padding: 2px"><span style="font-size:24px;"></span>'''Data Reference'''</div>= | ||
+ | |||
+ | * IS428 Visual Analytics for Business Intelligence week 1 to week 9 Lecture Notes | ||
+ | * Tableau Public | ||
+ | * DataCamp | ||
+ | * https://www.youtube.com/watch?v=OUTi12VtXpU | ||
+ | * https://www.youtube.com/watch?v=JodWmiIxl2c |
Latest revision as of 01:43, 18 March 2020
Contents
Project Motivation
Every two years, SMU Libraries conducts a comprehensive survey in which faculty, students and staff have the opportunity to rate various aspects of our services. The survey provides SMU Libraries with input to help enhance existing services and to anticipate emerging needs of SMU faculty, students and staff.
However, based on the current report provided by the library it requires readers to complete 155 pages of report in order to have a full understanding of the content. After reading through all the pages of the report, the report only provides a surface level of analysis in which the library cannot draw much insightful information out of it.
Thus, in order to have better visualization and understanding of the data from the survey results, I have decided to use Tableau as a platform which allows the Library to toggle between the various groups; Undergraduate, Postgraduate, Staff and Faculty and also offer several types of data visualization that can be more easily readable.
Project Objectives and task
In this assignment, the aim is to apply 10 Weeks of Visual Analytics for Business Intelligence module's knowledge with self-learning techniques to help Singapore Management University Library to deliver a focused and compact visualisation. By doing, it allows Singapore Management University Library to be well-informed of their service feedback from users as well as to identify those critical areas currently in the needs of improvement in order to satisfy user's experience.
Using visual analytics approach to reveal the level of services provided by SMU libraries as perceived by:
- the undergraduate students,
- the postgraduate students,
- the faculty,
- the staff.
Data Preparation
Data Cleaning Using Excel and Tableau
- Analysing the raw excel data, it becomes evident that there are multiple rows with data that are either an empty space " " or "NIL". Since both of them are referring to the same output. In order not to confuse the readers, I have decided to combine both data together and label it as "no input". Hence in ranking columns, it would show 1 to 7 and "no input".
- Under the question "Which library do you use most", the answer is reflected on the excel sheet as "1" or "2" where "1" is a representation of "Li Ka Shing Library" and "2" is a representation of "Kwa Geok Choo Library". This can be quite confusing, as the usage of numbers is to represent the ranking but it is only exceptional in this case. Therefore, I have decided to convert those relevant data into a word format instead of it being represented by numerical numbers.
Before |
---|
After |
- Tableau prefers data to be formatted in a machine-readable format. To reshape data for easier analysis in Tableau, I need to perform a pivot. In Tableau, pivoting means transposing data from a crosstab format into a columnar format–from wide, short tables into thin, tall tables. Tableau Prep Builder makes pivoting visual, allowing to see how data changes with every step.
- Connect to the library data source
- Drag the table that you want to pivot
- Click the down arrow icon and select Pivot from the context menu
- After selecting, the results appear immediately in the pivoted form
Before |
---|
After |
- After looking through the pivoted data and comparing with the report provided, I realised that the category types are missing in this data. These categories are important as it allows those relevant questions to group under the same category. So that it allows the users to choose how deep they prefer their data to be in terms of presenting the data visualization which allows the use not only limits to question along. With that, I will be extracting the latest pivoted data set from Tableau and saved as a csv file. On the csv file, I will create a new column and label it "Category". Under that column, I will insert the relevant grouping titles; Communication, Facilities and Equipment, Information Resources and Service Delivery. This is in accordance with the groupings as by the report provided.
Before |
---|
After |
Selection of Graphs used in Data Visualization
Graph Type | Reasoning | Type of Graph would like to use in library survey data analysis |
---|---|---|
Likert Scale |
|
|
KPI Performance Chart |
|
![]() Example of KPI Performance Chart Source:https://lh3.googleusercontent.com/-c8CKjPabp1g/V3FJWKo3KLI/AAAAAAAAyog/B8nHUvLW8tA/Screen-Shot-2016-06-27-at-4.31.26-pm.png?imgmax=9999 |
Vertical Bar Chart |
|
|
Horizontal Bar Chart |
|
![]() Example of Horizontal Bar Chart Source:https://peltiertech.com/images/2010-12/TextLabel2007Bar08.png |
Population Pyramid Chart |
|
|
Highlight Table |
|
|
Treemap |
|
![]() Example of Treemap Source:https://datavizcatalogue.com/methods/images/top_images/treemap.png |
Survey DashBoard Overview Information
Tableau Visualization
The interactive Tableua visualization can access here.
DashBoard Overview
DashBoard Summary
Picture Number | Description & Analysis |
---|---|
1 |
|
2 |
|
3 |
|
Introduction for Gap Analysis
Gap Analysis Explanation
- The highlight table shows the percentage breakdown of rating for both importance and performance for each group – Undergraduate, Postgraduate, Staff and Faculty.
- Since, the individual breakdown will be represented in the form of KIP Chart. For every KIP chart, there will be a KIP pre-set in order to conduct the comparison. In this case, the KIP will be the Importance figure set by the responders themselves which will be representing in the red line and the performance figure will be represented in the form of bar chart.
- Conducting a gap analysis where we take the Performance minus the Importance to find the gap between the importance of the question to them and compare it to how the library performs.
- Since Importance is defined as how important it is to that individual and performance is defined as how well the library is performing in that area.
- We will take the difference between these two, to measure how well the library is performing when compared to the level of importance held to them.
- If the gap analysis turns out positive, it means that performance has exceeded importance. If it turns out negative, it means that performance has not met the same ranking as importance. And this is one key aspect the library should look at when improving their services.
- I will be picking out 1 major change in each group for each category – Undergraduate, Postgraduate, Staff and Faculty and the rest can be viewed on Tableau Public.
Undergraduate Student
Gap Analysis
Picture Number | Description & Analysis |
---|---|
1 | Category: Communication
|
2 | Category: Service Delivery
|
3 | Facilities and Equipment
|
4 | Information Resource
|
Examining Correlation
- Using a highlight table to examine the correlation between library visits and campus visits, we can see the percentage breakdown of the student’s frequency.
- The table shows that although 25% indicated that they visit the school campus quarterly while also visiting the library daily.
- From the data shown, we can reach to a conclusion that it is likely impossible for a student to be visiting the campus quarterly and yet at the same time visiting the library daily. Hence, there might be a problem in the choices given to the students which led to this data given.
- So overall, this data is not accurate due to misinterpretation of the questions hence, this data cannot be used completely and the mistake only surfaced when the data is visualized.
Library Frequency Visits
- It can be seen from the breakdown of the library visits by undergraduates would be that majority of Business students patronize the library more than other schools. And the highest visits would be the weekly visits of business students across all schools at 42.44%.
Trend of Missing Data
- It is evident that under “Service Delivery”, the majority of Undergraduate students did not answer some of the questions in that section as compared to the rest. This is especially so for question 7 and question 9 where 12.09% and 9.63% did not answer the question.
- The questions that are generally left unanswered can be seen as irrelevant to the students as they probably never used that “service” or they do not require the library to provide them with such services.
Postgraduate Student
Gap Analysis
Picture Number | Description & Analysis |
---|---|
1 | Category: Communication
|
2 | Category: Service Delivery
|
3 | Facilities and Equipment
|
4 | Information Resource
|
Library Frequency Visits
- It can be seen that Postgraduates students that visit the library largely constitutes of business students. And the frequency of their visits is usually quarterly, at its highest of 64.71%.
- Whereas the percentage of visits, as well as the frequency of visits, is significantly lower for the other schools.
- The number of business students patronizing the library is almost double to triple the amount compared to other schools.
Trend of Missing Data
- Question 7 and Question 16 are seen to have the highest percentage of seeing no answers provided and again, this can be due to the nature of services or facilities which satisfies the needs of the target group.
- In the case of Postgraduates, it is evident that certain questions do not apply to them.
- For example, Question 16 is about “a computer is available when I need one”. This usually does not apply to Postgraduates as the majority would have had their own.
Faculty
Gap Analysis
Picture Number | Description & Analysis |
---|---|
1 | Category: Communication
|
2 | Category: Service Delivery
|
3 | Facilities and Equipment
|
4 | Information Resource
|
Library Frequency Visits
- From the data shown, we can see that the number of visits is evenly spread between 3 schools, namely; Accounting, Business and Economics.
- However, it can be seen that it is a consistent value of between 26%-29%.
- Law and Social Sciences do not even have the data for monthly, quarterly visits to the library, with the majority being daily and weekly.
Trend of Missing Data
- The high number of “no answers” is evidently flagged out under the “Facilities and Equipment ranging from 7-9%.
- With the second-highest coming from the category “Information Resources”, the highest at 5.34%.
- This shows that there is a large number of people under the Faculty category that did not answer these questions.
- As we can see those questions fall under “Facilities and Equipment” are generally asking about printing, photocopying facilities or can you find a quiet place for group project or study. As we know all the faculty member are able to get printing or photocopy facilities either from their own office or send for mass printing, which means that in most of the time they do not require such services. Not only that, but all the faculty members do also own an individual office to do their research or preparing for teaching material. Therefore, they do not require a place in the library.
Staff
Gap Analysis
Picture Number | Description & Analysis |
---|---|
1 | Category: Communication
|
2 | Category: Service Delivery
|
3 | Facilities and Equipment
|
4 | Information Resource
|
Data Reference
- IS428 Visual Analytics for Business Intelligence week 1 to week 9 Lecture Notes
- Tableau Public
- DataCamp
- https://www.youtube.com/watch?v=OUTi12VtXpU
- https://www.youtube.com/watch?v=JodWmiIxl2c