carl-alberto / hackbright

A repo with instructions and resources for the 4/20/17 Hackbright/Pantheon Challenge

Pantheon / HackBright Data Challenge

Schedule

• 6-6:30 Dinner, mingling
• 6:30-7:30 Group Problem Solving
• 7:30-8 Present as group

Problem Solving

In 2016, we organized a community initiative dubbed The Diversity Of The Web to survey, understand and celebrated diversity in the web development community. We got over 600 responses, about everything from gender identification to religious identification to disability identification.

Your task is to answer some questions from the result data, and prepare it for a 5-minute group presentation.

Here's what that might look like as a user-story:

As a member of the web community, I want insights into the diversity of the web-development community, so that I can better understand how to interact with and improve the community.

Resources

Join this Slack channel, for sharing links, requesting additional snacks, etc.

Here's how you can get the data:

• Live Survey
• Survey Questions
• Results as a CSV
• Results as SQL
• Results as Google Sheet
• Development environments on Pantheon with the data pre-loaded into MySQL (Ask Nick for a connection string)

Here's what we're looking for:

• Quantity/Difficulty: Can you answer all the questions? The difficult questions?
• Accuracy: Can you answer the questions accurately?
• Time/team management: Can you work as a group with a timing constraint? (hint: fewer answers beats no presentation)
• Presentation/Communication: Can you communicate the answers to your questions?
• Approach: How do you approach solving unfamiliar problems?

Groups

Each group has a few questions to answer. The questions generally start easy and get harder.

Group One

Spirit guides: Michelle & Ben

Answer these:

• For each role, what percentage of the community identifies as fullfilling that role?
• Is the data's ethnic breakdown roughly representative of the US's ethnic breakdown?
• Are community members from an affluent childhood more likely to have a technical degree?
• Are there any common answers among people who identify as other than M or F?
• (see common questions below)
• BONUS: What's another insight you had from the data?

Group Two

Spirit guides: Kim+Ian

Answer these:

• What percentages of the community live in urban/suburban/rural areas?
• Does the distribution of country of origin indicate an even sampling across continents?
• What roles were most often also formal managers?
• Is there a relationship between size of company and "at work I feel comfortable expressing identity"?
• (see common questions below)
• BONUS: What's another insight you had from the data?

Group Three

Spirit guides: Adam & Nikki

Answer these:

• What percentage of community members identified as a religious minority?
• Is the percentage of college graduates more or less than the average for the US?
• What roles were most often informal managers?
• What were the most common overlaps in role?
• (see common questions below)
• BONUS: What's another insight you had from the data?

Group Four

Spirit guides: Sujay & Kyle

Answer these:

• What percentage of community members are formal or informal managers?
• Is the data's gender breakdown representative of the US's gender breakdown?
• What are the urban/rural/suburban demographics of people who say "It's important to me to to feel comfortable?"
• Is there a relationship between gender and years of experience?
• (see common questions below)
• BONUS: What's another insight you had from the data?

Group Five

Spirit guides: Nick

Answer these:

• What percentage of the respondents are in each age grouping?
• Is the data's gender breakdown representative of the US's gender breakdown?
• Are people with physical disabilities more likely to live in urban areas?
• Is there a relationship between disabilities and company size?
• (see common questions below)
• BONUS: What's another insight you had from the data?

Common Questions

In one sentence…

• What biases could be represented in this analysis? How could we mitigate those biases?
• What effect could unanswered (optional) questions have on this analysis?