2b:null 2c:["$","title",null,{"children":"The Dual Effect of Music Therapy: Healing for Most, Harmful for a Few | Northeastern University John A. Guerra Gomez classes | Observable"}] 31:Td46,# The Dual Effect of Music Therapy: Healing for Most, Harmful for a Few This is a data visuzalization project made by [Shufan Chai](https://observablehq.com/user/@shufan-chai) from Northeastern University for the course CS7295: Information Visualization Fall 2024 with Professor John Alexis Guerra Gomez. Link to the course: https://johnguerra.co/classes/data_visualization_fall_2024/ ## Description Music therapy is a therapeutic approach that uses music to address physical, emotional, cognitive, and social needs of individuals. It can help improve mental health by reducing stress, enhancing mood, and promoting emotional well-being through various music-based interventions. You can click [here](https://en.wikipedia.org/wiki/Music_therapy) to learn more about Music therapy.

This project aimed to use a Parallel Coordinates Plot to visualize the relationships among individuals' mental health conditions, their hours spent listening to music and their mental health outcomes. The goal was to discover and summarize the mental health outcomes of individuals with various mental health conditions after listening to music for a certain number of hours per day. The four mental issues studied in this project are anxiety, depression, insomnia, and OCD.

Anxiety: A mental health condition characterized by excessive worry, fear, or nervousness that can interfere with daily activities.

Depression: A mood disorder marked by persistent feelings of sadness, hopelessness, and a lack of interest or pleasure in activities.

Insomnia: A sleep disorder that involves difficulty falling asleep, staying asleep, or getting restful sleep, despite having the opportunity to do so.

OCD: A condition in which a person experiences intrusive, unwanted thoughts (obsessions) and engages in repetitive behaviors or mental acts (compulsions) to reduce the anxiety caused by these thoughts.

## Table of Contents * [Part 1: (Real) Insights](https://observablehq.com/d/5b6924b271a09832#cell-17) * [Part 2: Data Abstraction](https://observablehq.com/d/5b6924b271a09832#cell-21) * [2.1 Data source](https://observablehq.com/d/5b6924b271a09832#cell-24) * [2.2 Dataset discription](https://observablehq.com/d/5b6924b271a09832#cell-27) * [2.3 Attributes used](https://observablehq.com/d/5b6924b271a09832#cell-30) * [2.4 Data availability](https://observablehq.com/d/5b6924b271a09832#cell-33) * [Part 3: Analysis Process](https://observablehq.com/d/5b6924b271a09832#cell-36) * [3.1 Example made-up insights and considerations](https://observablehq.com/d/5b6924b271a09832#cell-39) * [3.2 Target audience](https://observablehq.com/d/5b6924b271a09832#cell-42) * [3.3 Task abstraction](https://observablehq.com/d/5b6924b271a09832#cell-45) * [3.4 How abstraction](https://observablehq.com/d/5b6924b271a09832#cell-48) * [3.5 Data processing](https://observablehq.com/d/5b6924b271a09832#cell-51) * [Limitations and Future work](https://observablehq.com/d/5b6924b271a09832#cell-687) * [The D3 examples and API used in this project](https://observablehq.com/d/5b6924b271a09832#cell-95) * [tweet](https://x.com/cherryflaneur/status/1848221130198224913) & [video](https://youtu.be/YfVsFlu9Ebc) explaining the project32:T43b,

Most people experience emotional improvement from listening to music, especially those who listen to music for around 7-9 hours a day. The chart shows that the green lines are the most common, which means that many people with anxiety, depression, insomnia, and OCD feel that music improves their mental health. This is especially true when people listen to music for longer periods (around 7-9 hours per day). In general, music seems to help reduce stress and improve mood for most people.
A small number of people experience worsening symptoms from listening to music. Even though most people report positive effects, the purple lines show that a small number of people experience worse symptoms after listening to music. This happens more often in people who frequently experience or suffer from severe anxiety, depression, insomnia, or OCD, where the purple lines are easier to see. For these people, music might increase negative feelings, especially if they listen to it for shorter periods (less than 2 hours per day).

33:T737,{ // Reshape wide data to make it tidy. const points = dimensions.flatMap((dimension) => dataForChart1.map(({ [dimension]: value }, index) => ({ index, dimension, value })) ); // Compute normalization scales for each dimension. const scales = new Map( dimensions.map((dimension) => [ dimension, d3.scaleLinear().domain(d3.extent(dataForChart1, (d) => d[dimension])) ]) ); // Compute ticks for each dimension. const ticks = dimensions.flatMap((dimension) => { return scales .get(dimension) .ticks(7) .map((value) => ({ dimension, value })); }); return Plot.plot({ width: 2000, height: 1000, marginLeft: 104, marginRight: 20, x: { axis: null }, y: { padding: 0.1, domain: dimensions, label: null, tickPadding: 9}, color: { domain: ["Improve", "No effect", "Worsen"], range: ["green", "gray", "purple"], legend: true }, marks: [ Plot.ruleY(dimensions), Plot.lineX(points, { x: ({dimension, value}) => scales.get(dimension)(value), y: "dimension", z: "index", stroke: ({index}) => dataForChart1[index]["music_effects"], strokeWidth: ({index}) => { const effect = dataForChart1[index]["music_effects"]; return effect === "Worsen" ? 5 : effect === "No effect" ? 5 : 2; }, strokeOpacity: ({index}) => { const effect = dataForChart1[index]["music_effects"]; return effect === "Improve" ? 0.6 : effect === "No effect" ? 0.6 : 0.6; } }), Plot.text(ticks, { x: ({dimension, value}) => scales.get(dimension)(value), y: "dimension", text: "value", fill: "black", stroke: "white", strokeWidth: 3, fontSize: "24px" }) ] }); }34:T6d0,### 3.1 Example Made-up Insights and Considerations

Example Made-up Insight 1: Listening to music for 4 hours a day can significantly improve an individual's mental health, especially for those suffering from severe and persistent depression.
- Considerations:
  For me to test this, I need to have enough data from people with mental health conditions: the severity of their symptoms(Quantitative attribute), the number of hours they listen to music per day(Quantitative attribute) and the effects of music on them(Ordinal attribute).
  
  In terms of the idiom (visualization), I would need to display the hours each respondent's spent in music, and provide a way to demonstrate the effects of listening to music on different mental health contidions. The Parallel Coordinates Plot would be a good choice for this.
(selected) Example Made-up Insight 2: People who listen to music for around 7 hours a day tend to experience improvements in mental health symptoms.
- Considerations:
  For me to test this, I need to have sufficient data on hours_per_day and the associated music effects (ordinal attribute: Improve/No effect/Worsen). I would require data points from individuals who listen to music for 7-9 hours a day and their reported mental health outcomes (quantitative attributes: anxiety, depression, insomnia, OCD).
  
  In terms of the idiom (visualization), the same as above, I would need to display the hours each respondent's spent in music per day, and provide a way to demonstrate the effects of listening to music on different mental health contidions. The Parallel Coordinates Plot would be a good choice for this.

35:T83f,### 3.4 How Abstraction Mark:

Line: Each line represents a respondent data. These lines connect different axes, showing the how one respondent's data changes across the multiple dimensions.

Channels:

Position on common scale: Each quantitative attribute(hours_per_day, anxiety, depression, insomnia, and OCD) is arranged vertically along a common scale, forming clear separations. This uses position on a common scale as a magnitude channel to express relative relationships and value changes between attributes.
Position on unaligned scale: While each attribute has its own scale, the scales between different attributes are not necessarily aligned. The values of each attribute are displayed within their independent ranges, which corresponds to the use of an unaligned scale.
Color hue: The line's color represents the ordinal attribute: Music effects. Using color to represent this attribute helps easily distinguish the different outcomes of music therapy: Green for Improve; Gray for No effect; Purple for Worsen.

Expressiveness and Effectiveness:

Expressiveness: This plot is expressive because it uses lines to clearly show how each respondent's data point changes across multiple mental health dimensions. Position on common scale expresses the severity of each condition and the hours of listening to music. Position on unaligned scale expressively represents that each attribute has its own scale. Color hue clearly distinguishes the outcomes of music therapy (improve, no effect, worsen), making the relationships between attributes easy to interpret.
Effectiveness: By mapping the quantitative attributes (Anxiety, Depression, Insomnia, OCD, and Hours per day) vertically along a common scale, and the ordinal attribute (Music effects) to color, the plot effectively visualizes how different music effects (Improve, No effect, Worsen) relate to mental health conditions across respondents.

36:[{"id":"23803518ff9426bcaea13386a44a37515932c30b3f514328e0f83652590cad432467352a16ec0c83a6264cf7cb8cb05fea881425c1b3e8e19ad0a087a36adf88","href":"https://static.observableusercontent.com/files/23803518ff9426bcaea13386a44a37515932c30b3f514328e0f83652590cad432467352a16ec0c83a6264cf7cb8cb05fea881425c1b3e8e19ad0a087a36adf88","name":"mxmh_survey_results@2.csv","type":"text/csv","lastModified":1729312251575.927,"size":169089}] 2f:{"id":"5b6924b271a09832","slug":null,"title":"The Dual Effect of Music Therapy: Healing for Most, Harmful for a Few","thumbnail":"a84ad2618ccd5ac3d911d92fdd826b0f46bb53775d2997093d124989f5cdb348","access_level":"public","create_time":"$D2024-10-17T01:35:29.151Z","update_time":"$D2024-11-16T04:16:37.463Z","view_time":null,"first_public_time":"$D2024-10-17T21:35:48.399Z","has_unpublished_edits":null,"fork":false,"stars":0,"starred":false,"owner":{"type":"team","id":"afbaa6d30585fd36","login":"nujguerra","name":"Northeastern University John A. Guerra Gomez classes","avatar_url":"https://avatars-next.observableusercontent.com/04e5ef72b3edcda1296894b5d1ecaa9c92ef5d52d69e65e4c115de5534928fe9","tier":"education_2026","bio":""},"version":778,"publish_version":null,"role":null,"authored":false,"authors":[{"type":"individual","id":"a66d9c6f7c527559","login":"shufan-chai","name":"Shufan Chai","avatar_url":"https://avatars-next.observableusercontent.com/c9b39c358d361b518562d961d655caa1a16cee2580758c0c44efaf089197ceda","tier":null,"bio":"Thou art more lovely and more temperate."}],"members":[{"type":"individual","id":"a66d9c6f7c527559","login":"shufan-chai","name":"Shufan Chai","avatar_url":"https://avatars-next.observableusercontent.com/c9b39c358d361b518562d961d655caa1a16cee2580758c0c44efaf089197ceda","tier":null,"role":"editor"},{"type":"team","id":"afbaa6d30585fd36","login":"nujguerra","name":"Northeastern University John A. Guerra Gomez classes","avatar_url":"https://avatars-next.observableusercontent.com/04e5ef72b3edcda1296894b5d1ecaa9c92ef5d52d69e65e4c115de5534928fe9","tier":"education_2026","role":"viewer"}],"comments":[],"custom_thumbnail":true,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":null,"type":"notebook","body":{"titling":"auto","title":"The Dual Effect of Music Therapy: Healing for Most, Harmful for a Few","cells":[{"id":0,"value":"$31","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":17,"value":"## Part 1: Real Insights","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":561,"value":"$32","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":476,"value":"$33","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":21,"value":"## Part 2: Data Abstraction","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":24,"value":"### 2.1 Data Source\nThis dataset is from Kaggle and contains survey results on music taste and self-reported mental health. You can click [here](https://www.kaggle.com/datasets/catherinerasgaitis/mxmh-survey-results?resource=download) to view the data source.\n

The data collection was managed by [catherinerasgaitis](https://www.kaggle.com/catherinerasgaitis) via a Google Form. Respondents were not restricted by age or location.\nThe form was posted in various Reddit forums, Discord servers, and social media platforms. Posters and \"business cards\" were also used to advertise the form in libraries, parks, and other public locations.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":27,"value":"### 2.2 Dataset Discription\nThis dataset is a table that originally had 736 rows and 33 columns. For this project, I dropped all rows that contain null values.\n\nTherefore, the dataset I used for this project has 616 rows and 33 columns. Let's take a look at the first 10 rows of the dataset.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":101,"value":"printTable(cleanedData.slice(0, 10))","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":30,"value":"### 2.3 Attributes Used\n

Hours per day(Quantitative attribute): Number of hours the respondent listens to music per day.
Anxiety(Quantitative attribute): Self-reported anxiety, on a scale of 0-10. 0 means I do not experience this; 10 means I experience this regularly, constantly/or to an extreme.
Depression(Quantitative attribute): Self-reported depression, on a scale of 0-10, same as above.
Insomnia(Quantitative attribute): Self-reported insomnia, on a scale of 0-10, same as above.
OCD(Quantitative attribute): Self-reported OCD, on a scale of 0-10, same as above.
Music effects(Ordinal attribute): Does music improve/worsen respondent's mental health conditions? Improve/No effect/Worsen.

Psychology therapists who are interested in using music therapy to help individuals with mental illness.
Artists who care about social impact and would like to create music to support people with mental illness.

","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":45,"value":"### 3.3 Task Abstraction \nT1(Main task): Summarize the mental health outcomes of individuals with various mental health conditions after listening to music for a certain number of hours per day. (Discover/Browse/Summarize, Features)\n

T1.1(Secondary task): Compare the mental health outcomes of individuals who listen to more than 7 hours of music per day with those who listen for fewer hours. (Discover/Browse/Compare, Features)

","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":48,"value":"$35","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":51,"value":"### 3.5 Data Processing","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":199,"value":"#### Import data","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":92,"value":"data = FileAttachment(\"mxmh_survey_results@2.csv\").csv()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":381,"value":"#### Clean data\nI dropped all rows that contain null values.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":386,"value":"cleanedData = data.filter(d => {\n return Object.values(d).every(value => value !== null && value !== undefined && value !== \"\");\n})","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":450,"value":"#### Filter data for creating the Parallel Coordinates Plot(chart 1)","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":460,"value":"dataForChart1 = cleanedData.map(d => ({\n music_effects: d[\"Music_effects\"],\n hours_per_day: d[\"Hours_per_day\"], \n anxiety: d[\"Anxiety\"], \n depression: d[\"Depression\"], \n insomnia: d[\"Insomnia\"], \n ocd: d[\"OCD\"] \n}))","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":464,"value":"dimensions = Object.keys(dataForChart1[0]).slice(1);","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":234,"value":"#### Overview of data using Navio\nNavio is a data exploration widget created by [Professor John Alexis Guerra Gomez](https://observablehq.com/@john-guerra/navio), which you can use to explore the dataset.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":242,"value":"navio(cleanedData, { attribWidth: 27 })","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":687,"value":"## Limitations and Future work\nFirst, in this project, the music effects data comes from self-reports by the participants, reflecting their subjective feelings. It lacks clinical rigor, so the conclusions of this project also lack clinical precision to some extent and should only be considered as preliminary research references. \n

Second, this project only focused on the time spent listening to music without further examining specific music types. Therefore, in the future, the project can delve deeper into exploring the relationship between music types, participant age, and other variables with music effects to draw more comprehensive insights.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":95,"value":"## The D3 examples and API used in this project","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":97,"value":"import {printTable} from \"@uwdata/data-utilities\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":201,"value":"import { vl } from \"@vega/vega-lite-api-v5\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":236,"value":"import {navio} from \"@john-guerra/navio\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"}],"files":"$W36","stdlib":"1"}} 30:"$2f:body"