sfbrigade / analytics Goto Github PK

Integrate the Airtable New Member Form into pipeline. Ensure it's fully automated.

Current export via pandas leaves null in column 1, the unique key column from the DataFrame.

Please review the member summary code for any pandas/python improvements.

Compare proportion of public posts by male vs female in project by gender of project lead.

  | M | F.  <-- Project Leads

M. |. |
F. |. |
^ Participants

Take from conversations.list for channel name, conversation history for number of public messages, and member form for genders; create crosstab.

project channels with female leads:
discuss-data-science
proj-datasci-bay-releif
proj-waterthetrees

Mix leads
proj-intentional-walk (mix)
proj-teambats

Project channels with male leads:
issue-covid-19
proj-3d-street
proj-compass
proj-tinytown
proj-brigade-analytics
proj-brigade-website
proj-police-budgets

Brigade Channels
general
discuss-new-member
share-civic-jobs
share-events

Metric is count of posts by gender

https://api.slack.com/methods/conversations.history

Create dump of messages with each channel name and user name attached.

Getting a Python error when trying to run latest version of users_list.py.

The file itself seems to have a combination of stashed/local development changes.

Many usernames and emails are displaying in member_analytics as NaN. When queuing user_list_data (https://api.slack.com/methods/users.list), example id 'U02BF7TJ99V' has neither value populated. However, when exporting the user list from slack admin tools, I see "Cultivate Team" as the name, and an email is present.

Create a data set for user.list

Add activity buckets so we can build a small histogram for member activity. This will help to see how member engagement shapes up over time..

These are a signal of activity

To help with more targeted out reach

One question that comes here is in terms of percent of slack activity, how much can be associated with demographic data?

https://api.slack.com/methods/calls.info

Get calls, dates, users, channels into a single file

Automate the pipeline so that it can run in one go

This is the list of channels in slack. Include archived.

https://api.slack.com/methods/conversations.members

Includes a list of members for each channel. How do we use this?

Does joining a secondary channel or being responded to have a higher impact on retention?

https://api.slack.com/methods/team.accessLogs

For each user in users.list, add the following to the data set.

first access
most recent access
number of logins
country
region

Note, does not have paging, need to increment through page until all results are found.

Load entire log, identify the aggregate values based on userid, then join to users.list info

Validate that email address can be used to join Airtable Demographic data with slack data.

Answer the following questions:

1. How many users are in the airtable?
2. How many users are in the slack list?
3. How many users are in both the airtable and slack?
4. For those who are in list 1, but not in list 2, How many can we match manually if we were to use name and not email addresses?