2b:null 2c:["$","title",null,{"children":"Low salaried, married graduates use the most number of credit cards! | Zhenghao Lin | Observable"}] 31:T52e,{ const hover = vl.selectMulti().fields(["Card_Category"]); const cardHistogram = vl .markBar({tooltip: true}) .select(hover) .encode( vl.y().fieldN("Card_Category").sort("-x"), vl.x().count().title("Count of individuals per card category"), vl.color().if(hover, vl.color().value("firebrick")).value("#3333"), ) .height(200) .width(500); const cardByAgeGroup = vl .markSquare({tooltip: true}) .transform(vl.filter(hover)) .encode( vl.x().fieldN("Income_Category").sort(["Less than $40K", "$40K - $60K", "$60K - $80K", "$80K - $120K", "$120K +"]), vl.y().count().title("Count of male/female"), vl.size().count(), vl.color().fieldN("Gender"), vl.tooltip([vl.y().count()]) ) .height(400) .width(270); const cardByMaritalStatus = cardByAgeGroup.encode( vl.x().fieldN("Marital_Status").sort(["Single", "Married", "Divorced"]) ); const cardByEducationLevel = cardByMaritalStatus.encode(vl.x().fieldN("Education_Level").sort(["Unedcated", "High School", "College", "Graduate", "Post-Graduate", "Doctorate"])); const bubblePlots = vl.concat( cardByAgeGroup, cardByMaritalStatus, cardByEducationLevel ); return vl .vconcat(cardHistogram, bubblePlots) .data(data) .render(); }32:T1577,- **Hypothesis** : Gold or Platinum should be the most used credit card type by individuals who are married, earn more than 80K annually and have at least graduation degree. - **Consideration** : I will need enough credit card holders data that entails the type of credit card used by holder, their income range, level of education and also their marital status. - **Main Insights** : Out of the total cardholders, only 11 have platinum cards while a significant 6,598 individuals hold blue credit cards. The majority of blue cardholders, who are typically married and hold a graduation degree, earn an annual income of less than 40k. - **Tasks (Why)** : - **T1** : Compare the number individuals holding blue, silver, gold and platinum credit cards **[Compare, Feature]**. - **T2** : Summarize the distribution of men and women by income category, marital status and education level **[Summarize, Distribute]**. - **T3** : Identify the income range and number of men and women in that range **[Identify, Features]**. - **T4** : Identify the marital status and number of men and women married, single or divorced **[Identify, Features]**. - **T5** : Identify the education level and number of men and women who have completed different levels of education **[Identify, Features]**. - **T6** : identify the gender of credit card holders **[Identify, Feature]**. - **Data Abstraction (What?)** : * **Dataset type:** Tabular * **Attributes used:** * Card_Category * Attribute Type - Categorical * Income_Category * Attribute type - Categorical * Marital_Status * Attribute type - Categorical, * Education_Level * Attribute type - Categorical * Gender * Attribute type - Categorical * **Derived Attributes:** * "Count of individuals per card category" which is an aggregation of number of individuals grouped by the credit card type they use. * Attribute Type - Quantitative * Orientation - Sequential * "Count of Male & female" which is an aggregation of number of males and females first grouped by credit card type and second by income category/ marital status/ education level. * Attribute Type - Quantitative * Orientation - Sequential * **Availability:** Static * **Source:** [Kaggle - Bank Churners Data](https://www.kaggle.com/datasets/manjuahuja/bank-churner) - **Marks & Channels (How?)** : * The data is represented using a 2-D Map. * **Marks:** Points, Lines * For **Bar Graph**, the **x-axis** represents credit card type which are **separate, ordered and align** and the **y-axis** represents the total number of card holders which is **express**. * For the 3 **Scatter Plots**, the **x-axis** represents income_category, marital_status and education_level respectively which are **separate and aligned**. The **y-axis** for all 3 scatter plots represent the number of male and female, which is **express**. * **Channels:** - **Color Hue (Scatter Plot)** : To identify the gender of credit card holders. * Attribute : Gender **(Separate, Aligned)** * Task : T6 * Representing gender, a categorical attribute through the use of distinct colors on the scatter plot makes the visualization more **effective** by facilitating resource identification. Using color hue to differentiate between resources enhances the **expressiveness** of the channel, making it easier for the viewer to quickly understand the information being presented. The use of color in this manner greatly enhances the interpret-ability of the visualization and enables the user to gain a more comprehensive understanding of the distribution of men ad women. - **Size & Position (Scatter Plot)** : To identify the number of men and women across different characteristics like their marital status, education level and income category. * Attribute : Count of Male & female **(Express)** * Task : T3, T4, T5 * The count of Male & female, a quantitative attribute, for each unique characteristic can be **effectively** represented using points of varying sizes as well as their position with respect to y-axis in the visualization. This approach provides a concise yet informative representation of the total number of men and women. The use of different point sizes enables the viewer to quickly understand and quantify the distribution, making the visualization more **expressive** and easier to understand. - **Length (Bar Graph)** : To compare and identify the number of individuals using different credit cards (blue. silver, gold & platinum). * Attribute : Count of individuals per card category **(Express)** * Task : T1 * Count of individuals per card category, being a quantitative attribute, is best represented using the length channel in the visualization. The use of length allows for the clear expression and comparison of the count of individuals of different typr of credit card users, making it an effective channel for the given task. The ability to visualize and compare the the total number of individuals through length enhances the interpret-ability of the visualization and makes it easier for the viewer to understand and evaluate the information being presented. The **effectiveness** of this channel in representing the total number of individuals makes it an important tool for providing a clear and **expressive** representation of individuals and their preferred credit card type. 33:[{"id":"c7cfea4a76a71bc0428d63b3b3e506faa2572776e2a3376aa2fd269d2385da8a506e9d46167293aab81e5733ba1bc73ff709186a642b8b5d785a473db5f56354","href":"https://static.observableusercontent.com/files/c7cfea4a76a71bc0428d63b3b3e506faa2572776e2a3376aa2fd269d2385da8a506e9d46167293aab81e5733ba1bc73ff709186a642b8b5d785a473db5f56354","name":"BankChurners.csv","type":"text/csv","lastModified":1677700659286.192,"size":1510880}] 2f:{"id":"da118a54cc06439e","slug":null,"title":"Low salaried, married graduates use the most number of credit cards!","thumbnail":"df6ae557eeaf72b74c0e271cd7b7c12b11b56d148f000f65c1e54b927fc1578b","access_level":"public","create_time":"$D2023-03-01T19:57:39.286Z","update_time":"$D2023-03-01T20:22:29.040Z","view_time":null,"first_public_time":"$D2023-03-01T19:57:39.286Z","has_unpublished_edits":null,"fork":{"id":"52568ad264d76972","slug":null,"title":"Low salaried, married graduates use the most number of credit cards!","owner":{"id":"29c4858a5f0f02f2","login":"parthivi-varshney","name":""}},"stars":0,"starred":false,"owner":{"type":"team","id":"e7d220f596a8e193","login":"zhenghao-lin","name":"Zhenghao Lin","avatar_url":"https://avatars-next.observableusercontent.com/03a225dc3cc6dbe00ba43fafe9901db25cafa64a81b5555e001748a80c4b32d8","tier":"starter_2024"},"version":568,"publish_version":null,"role":null,"authored":false,"authors":[{"type":"individual","id":"4526580224c72b33","login":"zhenghao-lin","name":"Zhenghao Lin","avatar_url":"https://avatars-next.observableusercontent.com/03a225dc3cc6dbe00ba43fafe9901db25cafa64a81b5555e001748a80c4b32d8","tier":null}],"members":[{"type":"individual","id":"4526580224c72b33","login":"zhenghao-lin","name":"Zhenghao Lin","avatar_url":"https://avatars-next.observableusercontent.com/03a225dc3cc6dbe00ba43fafe9901db25cafa64a81b5555e001748a80c4b32d8","tier":null,"role":"editor"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":null,"type":"notebook","body":{"title":"Low salaried, married graduates use the most number of credit cards!","cells":[{"id":213,"value":"# Low salaried, married graduates use the most number of credit cards!\nAuthor : [Parthivi Varshney](https://observablehq.com/@pv1995)
\nCourse : [Data Visualization / Information Visualization](https://johnguerra.co/classes/data_visualization_spring_2023/)
\nInstructor : [John Alexis Guerra Gomez](https://observablehq.com/@john-guerra)
\n\n---","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":567,"value":"# I found your study very interesting about which group of people use the credit card the most. But I found that you did not take into account the proportion of a group of people in all the population. For example, for your marital status analysis, you found the divorced people use the least credit cards. However, it is most likely that divorced people is the least of all. I think you need to take into account how many of the all population are divorced, single, or married.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":405,"value":"### Introduction
\nA credit card is a payment card issued to users (card holders) to enable the cardholder to pay a merchant for goods and services based on the card holder's accrued debt. This notebook explores the [Bank Churners](https://www.kaggle.com/datasets/manjuahuja/bank-churner) data and sheds insight on the credit card usage of card holders among different demographic groups including age, income, marital status, education level and gender.
\n**[Click to watch the Video explanation](https://www.youtube.com/watch?v=c6KlOe62CMg)**
\n**[Click to read the Tweet](https://twitter.com/parthivivarshn1/status/1628292853473607680?s=46&t=txVTMMebunfpArzcMNJhGQ)**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":432,"value":"---","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":495,"value":"### Visualization using 5 attributes.\nThe graph is interactive, you can select one or multiple card categories.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":308,"value":"$31","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":358,"value":"$32","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":388,"value":"---","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":415,"value":"### Graph 1","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":13,"value":"viewof stackedBar = vl.markBar({ tooltip: true })\n .data(data)\n .encode(\n // vl.column().fieldO(\"Customer_Age\"),\n vl.x().fieldN(\"Income_Category\").axis({ title: \"Income Range\" }).sort([\"Less than $40K\", \"$40K - $60K\", \"$60K - $80K\", \"$80K - $120K\", \"$120K +\"]),\n vl.y().count(),\n vl.color().fieldN(\"Age_Group\").scale({ range: \"category\" }),\n )\n .width(400)\n .height(500)\n .render()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":247,"value":"- **Hypothesis** : Individuals earning at least 80K annually within the age group of 36-45 should hold most number of credit cards.\n- **Insight 1** : Most credit card holders earn less than 40k annually and are aged between 45-56 years.\n- **Tasks** : \n - **T1** : Identify the income range of credit card holders **[Identify, Feature]**. \n - **T2** : Identify the age group of credit card holders **[Identify, Feature]**.\n - **T3** : Compare the total number of credit card holders distributed amongst different income category **[Compare, Distribution]**.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":391,"value":"---","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":418,"value":"### Graph 2","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":53,"value":"viewof groupStackedBar = vl.markLine({point: true, tooltip: true })\n .data(data)\n .encode(\n vl.x().fieldN(\"Age_Group\").axis({ title: \"Age Groups\" }),\n vl.y().count(),\n vl.color().fieldN(\"Gender\").scale({ range: \"category\" }),\n )\n .width(400)\n .render()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":257,"value":"- **Hypothesis** : Men tend to use more credit cards than women.\n- **Insight 2** : Of all the credit card holders between the age of 46-55, there are more male than female.\n- **Task** :\n - **T1** : Identify the gender of credit card holders **[Identify, Feature]**.\n - **T2** : Compare the number of male and female in each age group **[Compare, Feature]**.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":394,"value":"---","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":420,"value":"### Graph 3","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":158,"value":"viewof lineGraph = vl.markRule({point: true, tooltip: true })\n .data(data)\n .encode(\n vl.y().fieldN(\"Income_Category\").axis({ title: \"Income Category\" }).sort([\"Less than $40K\", \"$40K - $60K\", \"$60K - $80K\", \"$80K - $120K\", \"$120K +\"]),\n vl.x().count(),\n vl.color().fieldN(\"Marital_Status\").scale({ range: \"category\" }),\n )\n .height(400)\n .width(400)\n .render()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":282,"value":"- **Hypothesis** : Since bachelors are more spendthrift, single individuals should have more credit cards than married or divorced individuals.\n- **Insight 3** : Among people of different marital statuses, divorced individuals used the least credit cards where as married individuals used the most number of credit cards across all income ranges.\n- **Task** :\n - **T1** : Identify the marital status of the card holders **[Identify, Feature]**.\n - **T2** : Compare the trends in the number of credit card holders across various income categories. **[Compare, Trend]**.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":397,"value":"---","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":422,"value":"### Graph 4","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":287,"value":"viewof pointGraph = vl.markBar({tooltip: true })\n .data(data)\n .encode(\n vl.x().fieldN(\"Income_Category\").sort([\"Less than $40K\", \"$40K - $60K\", \"$60K - $80K\", \"$80K - $120K\", \"$120K +\"]),\n vl.y().count(),\n vl.size().count(),\n vl.color().fieldN(\"Education_Level\")\n )\n .height(500)\n .width(400)\n .render()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":294,"value":"- **Hypothesis** : Since individuals with higher education level earn more money, they should hold more credit cards. So accordingly, the number doctorate holders and post graduates who use credit cards should be more.\n- **Insight 4** : Graduates earning less than 40k annually hold the most number of credit cards, while post-graduates earning more than 120k annually have the least number of credit cards.\n- **Tasks** :\n - **T1** : Identify the number of credit card holders, their education level and their annual income range **[Identify, Feature]**.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":400,"value":"---","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":407,"value":"### Audience\n- Bank executives: to help decision-making within the bank, bank executives and managers would be the primary target audience. They may be interested in seeing which factors are most strongly correlated with customer churn, such as customer demographics. They may also want to see which customer segments are most likely to churn, so they can develop targeted retention strategies.\n- Data analysts and scientists: The data analysts and scientists would be another primary target audience. They may be interested in exploring the data to identify patterns and relationships that are not immediately obvious, or in building predictive models to forecast future customer churn.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":426,"value":"---","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":424,"value":"### Visual Exploration Process\n- Graph 1 : Groups all the credit card users by their annual income range and age groups, to determine which age group and which income category individuals used most number of credit cars.\n- Graph 2 : Groups all these individuals in different age groups by their gender, to determine whether male used more credit cards or females.\n- Graph 3 : Shed a light on the marital status of most credit card users by grouping all the users by their marital status and also represents what income category they fall under.\n- Graph 4 : Summarizes and identifies the total number of individuals belonging to different income categories by their education level.\n- The main visualization provides a deeper look into these categories (income range, marital status and education level) by grouping the credit card users based on the type of credit card they used.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":409,"value":"---","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":411,"value":"### Data Manipulation\nThe Data manipulation is done using JavaScript. I first filtered the data to remove rows that contained \"Unknown\" value for income category, marital status and education level. Second I added a key in each object called \"Age_Group\", that grouped the data in defined age intervals. This filtered new data is served to vegalite for creating the above visualizations. ","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":413,"value":"---","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":98,"value":"function groupAges(item){\n let ageGroup = \"\"\n if(item.Customer_Age > 25 && item.Customer_Age <= 35){\n ageGroup = \"26-35\"\n }\n else if(item.Customer_Age > 35 && item.Customer_Age <= 45){\n ageGroup = \"36-45\"\n }\n else if(item.Customer_Age > 45 && item.Customer_Age <= 55){\n ageGroup = \"46-55\"\n }\n else if(item.Customer_Age > 55 && item.Customer_Age <= 65){\n ageGroup = \"56-65\"\n }\n else if(item.Customer_Age > 65 && item.Customer_Age < 74){\n ageGroup = \"66-73\"\n }\n return {...item, Age_Group:ageGroup}\n} ","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":102,"value":"data = filteredData.map(groupAges)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":68,"value":"filteredData = rawData.filter(d => d.Income_Category != \"Unknown\" && d.Marital_Status != \"Unknown\" && d.Education_Level != \"Unknown\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":0,"value":"rawData = FileAttachment(\"BankChurners.csv\").csv({ typed: true })","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":8,"value":"import { vl } from \"@vega/vega-lite-api-v5\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"}],"files":"$W33","stdlib":"1","titling":"auto"}} 30:"$2f:body"