Mapa mental

Data Analysis Checklist

Narendran Antony-Pillai Inglês

Conteúdo do mapa

1. Ask
- Ask effective questions
  - 5. Define what would the project look like
  - 6. Define what would qualify as a successful result
  - 7. Ask stakeholders what data to collect
    - 8. Make sure you fully understand the stakeholder's expectations
      - 12. Create a Scope Of Work document and fill in the document
        A Scope Of Work document requires the following information:
        Deliverables: What work is being done, and what things are being created as a result of this project? When the project is complete, what are you expected to deliver to the stakeholders? Be specific here. Will you collect data for this project? How much, or for how long?
        Milestones: This is closely related to your timeline. What are the major milestones for progress in your project? How do you know when a given part of the project is considered complete?
        Timeline: Your timeline will be closely tied to the milestones you create for your project. The timeline is a way of mapping expectations for how long each step of the process should take. The timeline should be specific enough to help all involved decide if a project is on schedule. When will the deliverables be completed? How long do you expect the project will take to complete? If all goes as planned, how long do you expect each component of the project will take? When can we expect to reach each milestone?
        Reports: Good SOWs also set boundaries for how and when you’ll give status updates to stakeholders. How will you communicate progress with stakeholders and sponsors, and how often? Will progress be reported weekly? Monthly? When milestones are completed? What information will status reports contain?
        Purpose
        Major Project Activities
        This project does not include
        Estimated date for completion
        Remember to add a deadline for the data cleaning process
        13. Clarify requirements of the project with stakeholders
        11. Know your project: Keep track of your discussions about the project over email or reports, and be ready to answer questions about how certain aspects are important for your organization. Get to know how your project connects to the rest of the company and get involved in providing the most insight possible. If you have a good understanding about why you are doing an analysis, it can help you connect your work with other goals and be more effective at solving larger problems.
      - 9. What are the boundaries for the project?
        10. What do you consider "in-scope"?
        11. Is there any other information you think is relevant to the project?
  - 4. Define the problem or question to answer
  - Communicate with others
  - 3. Who are the relevant sponsors and stakeholders for this project? Who is involved, and how?
  - 1. Align your file naming with your team's or company's existing file-naming conventions
    - Ensure that your file names are meaningful
      - Include information like project name and anything else that will help you quickly identify (and use) the file for the right purpose
        Include the date and version number in file names, common formats are YYYYMMDD for dates and v## for versions (or revisions)
        Create a text file as a sample file with content that describes (break)
        Avoid spaces and special characters in files names. Instead use dashes, underscores, or capital letters
      - Stick to a consistent pattern that contains the most useful information needed
        Keep file names short and to the point
    - Best practices for file naming conventions:
    - 2. Keep files organised
      - Create folders and subfolders in a logical hierarchy so related files are stored together
        Separate ongoing from completed work so your current project files are easier to find.
        Archive older files in a separate folder, or in an external storage location
        If your files aren't automatically backed up, manually back them up often to avoid losing important work.
2. Prepare
- 2. Build project timeline
  - 1. Identify what I need to figure out how to solve this problem
  - Remember to add a deadline for the data cleaning process
- 3. Identify what data is needed to achieve the successful result identified in the previous step
  - 4. Decide what specific information would be gathered
- 5. Establish rules for who would have access to the data collected
  - 6. Decide how will the data be managed
  - 7. Decide who will be responsible for the data
  - 8. Decide how and where the data is stored and tools used to do so
    - 9. Decide how the data will be protected
- 10. Decide if data collection is required and if required, decide how you will collect data and how much to collect
  - If required collect data
    - If you are collecting your own data, decide how long you will need to collect it
  - If creating a survey, ask for permission from surveyees before conducting the survey
    - Make sure the survey is inclusive
    - Make sure the surveyees understand how their data would be collected, stored, managed and protected
- 11. Decide how to best present the data visually
- 12. Brainstorm possible project and data related issues and how to avoid them
- 13. Store the data
  - 14. Understand how data is generated and collected
    - 15. Make sure that the data is unbiased and credible
      - 16. Protect the data
3. Process
- 2. Create a changelog and write down all data transformation and cleaning actions in that document
  - 6. Clean the data to make sure it is complete, correct, relevant and consistent
    - 7. Upload raw data to an internal data warehouse for an additional layer of security
      - 8. Contextualize the data
        Answer the following questions to contextualize the data:
        Who: The person or organization that created, collected, and/or funded the data collection
        What: The things in the world that data could have an impact on
        Where: The origin of the data
        When: The time when the data was created or collected
        Why: The motivation behind the creation or collection
        How: The method used to create or collect it
        9. Inspect the data to determine if it contains the specific information you need to answer your stakeholders questions
    - Make sure the data is free of outliers, errors, inaccuracies, inconsistencies, missing, faulty and outdated data
      - If using SQL, use SQL functions to check for extra spaces
        Remove repeated entries
        Double check for spelling errors and fix spelling if there are spelling errors in the dataset
        Check for misfielded values in the dataset and fix them if they occur in the dataset
        Change the case of text
        To change the case of text in Microsoft Excel:
        To convert text to proper case in Excel:
        PROPER function in Excel meaning: Capitalizes the first letter in a text string and any other letters in text that follow any character other than a letter. Converts all other letters to lowercase letters.
        Use the formula "=PROPER(CELL LETTER AND NUMBER YOU WANT TO CONVERT)" to convert the cell to proper case
        To convert text to lowercase in Excel:
        Use the formula "=LOWER(CELL LETTER AND NUMBER YOU WANT TO CONVERT)" to convert the cell to lower case
        To convert text to uppercase in Excel:
        Use the formula "=UPPER(CELL LETTER AND NUMBER YOU WANT TO CONVERT)" to convert the cell to upper case
        To change the case of text in Google Sheets:
        Click on the Add-Ons option at the top of Google Sheets.
        Click on Get add-ons.
        Search for ChangeCase
        Click on Install to install the add-on
        Highlight the cells you want to change
        Click on the Add-Ons tab
        Select ChangeCase
        Select the option you wish to change to
        Convert numbers stored as text to numbers
        To convert numbers stored as text to numbers in Excel:
        Select the cell that needs converting and click on the exclamation mark to choose a convert option
        Combine mutliple columns into a single column
        To combine mutliple columns into a single column in Excel:
        Use the formula "=CONCAT(CELL LETTER AND NUMBER YOU WANT TO COMBINE,CELL LETTER AND NUMBER YOU WANT TO COMBINE)"
        Split single column text into multiple columns
        To split single column text into multiple columns in Excel:
        Select the cell or column that contains the text you want to split.
        Select Data > Text to Columns.
        In the Convert Text to Columns Wizard, select Delimited > Next.
        Select the Delimiters for your data. For example, Comma and Space. You can see a preview of your data in the Data preview window.
        Select Next.
        Select the Destination in your worksheet which is where you want the split data to appear.
        Select Finish.
        To split single column text into multiple columns in Google Sheets:
        Type "=SPLIT(CELL LETTER AND NUMBER, SPLIT BY X CHARACTER". For example: If cell F2 had the info "A-B", the split function would separate A and B by the - (minus) symbol. So in this example, the function would be =SPLIT(F2,"-"). So when you execute this function in cell G2, you would get A on cell G2 and B on cell H2.
        Remove extra spaces
        In Google Sheets:
        Select the data cell or cells that you want to remove extra spaces in.
        Click on the Data tab, then hover over Data cleanup and select Trim whitespace.
        In SQL:
        UPDATE DATASET NAME.TABLE NAME SET COLUMN NAME = TRIM(COLUMN NAME) WHERE TRUE;
        Select and remove rows containing blank cells
        In Google Sheets and Excel:
        Highlight all cells in the spreadsheet
        Click on the Data tab and pick the Create a filter option. In Microsoft Excel, this is called Filter.
        Every column now shows a green triangle in the first row next to the column title. Click the green triangle in a column that you would like to check for blank cells, this will open a new menu
        On that new menu, click Filter by condition and open the dropdown menu to select Is empty. Click OK.
        In Excel, click the dropdown, then Filter... then make sure only (Blanks) is checked. Click OK.
        You can then review a list of all the rows with blank cells in that column.
        Select the row that has the blank cells in them by right clicking the row number and clicking the delete row button in the menu to delete the row with the blank cells in them
        To convert positive numbers to negative numbers
        In Google Sheets:
        Select an empty cell and type "=COLUMN NUMBER TO CONVERT*-1". For example if you wanted to convert the cell G2 from a positive number to a negative number, type ="G2*-1"
        If there is a data description, check if the dataset matches with the data description.
        To do this in SQL:
        For variable checking:
        Query the following: SELECT DISTINCT VARIABLE NAME TO CHECK FROM DATASET NAME.TABLE NAME;
        For range checking:
        Query the following: SELECT MIN(VARIABLE NAME) AS min_length, MAX(VARIABLE NAME) AS max_length FROM DATASET NAME.TABLE NAME;
        Check to see if you have missing data
        To do this in SQL:
        SELECT * FROM DATASET NAME.TABLE NAME WHERE VARIABLE NAME TO CHECK IS NULL;
        Fix misspelt data
        To do this in SQL:
        Write the following query and execute it: UPDATE DATASET NAME.TABLE NAME SET NAME OF COLUMN = "CORRECT WORD" WHERE COLUMN NAME = "WRONG WORD";
        Length checking
        To do this in SQL:
        Query the following: SELECT DISTINCT COLUMN NAME, LENGTH(COLUMN NAME) AS string_length FROM DATASET NAME.TABLE NAME;
        Make sure that the date formats are consistent throughout the dataset
        Make sure that the columns names are meaningful
        Review the goal of your project
        Confirm the business problem
        Confirm the goal of the project
        Verify that data can solve the problem and is aligned to the goal
        In Microsoft Excel:
        1. Select the range of cells, or ensure that the active cell is in a table.
        2. On the Data tab, click Remove Duplicates (in the Data Tools group).
        In Google Sheets
        Select the data range that you want to remove duplicate data in.
        Click Data and then Remove duplicates.
        Select which columns to include and whether the data has headers.
        Click Remove duplicates.
        In the status window, click OK.
      - If there is Null data in the dataset:
        Understand how Null was used before you start analyzing a dataset with Null data. Sometimes Null can be interpreted as the value, 0
        If there is outdated data in the dataset, try to replace that data with newer and more accurate data
      - Set up spell check in the program you are using to spell check the data, for example: Microsoft Excel or Google Sheets
        To run the spelling checker in Microsoft Excel, press F7 on the keyboard
  - 3. Organise and structure the data into a table so that it is useful
    - 5. Make sure to document errors
    - 4 Create a backup of the dataset before cleaning the data
  - Typically, a changelog records this type of information: Data, file, formula, query, or any other component that changed, Description of what changed, Date of the change, Person who made the change, Person who approved the change, Version number, Reason for the change
- 1. Restrict access to the collected or gathered data to a limited number of analysts
4. Analyse
- 1. Sort, filter and format the data so that you see the data you need for analysis
  - 3. Analyse the clean data to find insights and solve the problem/answer the question
    - 4. Document exactly what you found in the analysis, no matter what the results
      - 5. Make recommendations
        6. Make data-driven decisions
        7. What story is my data telling me?
        8. How will my data help me solve this problem?
    - Look for patterns, relationships and trends in the data
  - 2. If you are confused about the data in the dataset, contact the owner of the dataset for clarification
5. Share
- 6. Share the data with your audience
  - Communicate to help others understand results
- 2. Create effective visuals
  - 3. Use data storytelling
    - Summarise your results
    - 4. How can I make what I present to the stakeholders engaging?
      - 5. What would help me understand this if I were the listener?
  - Make visual descriptive
    - Make data easy to understand
      - If you need to show a change of values over time, line charts or bar graphs might be the best choice
        If your goal is to show how each part contributes to the whole amount being reported, a pie or donut chart is probably a better choice
        Keep your audience in mind
- 1. Choose which software you will use to visualise data, for example: Spreadsheets, Tableau, R
6. Act
- 1. Work with stakeholders to decide how to best implement changes and take actions based on the findings
  - 2. Keep relevant data stored for long-term and future reference
    - 3. Remove data from storage and delete any shared copies of the data
  - Provide stakeholders with recommendations based on your findings

Informações

Descrição: Mapa Mental sobre Data Analysis Checklist, criado por Narendran Antony-Pillai em 21-11-2022.
Criado por: Narendran Antony-Pillai
Visualizações: 0
Avaliação

Data Analysis Checklist

Conteúdo do mapa

Informações

Fale Conosco

O que você acha do novo visual?