> For the complete documentation index, see [llms.txt](https://gosleep.gitbook.io/white-paper/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://gosleep.gitbook.io/white-paper/sleep-assistant/data-analysis.md).
# Data Analysis
Dig inside to know what's going on.
We collect data from user and put it into analysis. Digging from real data leads us to the truth of sleeping.
There're already mature methods to learn from sleeping data. We follow well known and proven sleep data analysis patterns scientifically.
## Daily Hours Slept
Daily analysis allows you to see the variability in sleep duration over time
## Weekly Average Hours Slept
When averaging sleep data by week, you lose variability and there less to see…
## Bed Time Start vs. Average Hours Slept
As bed gets later (x-axis), hours slept get shorter (y-axis).
## Weekly Average Hours Slept vs. Weekly Average Steps
Maybe this is a fatigue factor? Correlation? But convincing evidence nonetheless. As this client became more consistent with physical activity (last 4 weeks), average hours slept increased as well.
## Hours Slept By Bed Time Start with Trend Line
A scatterplot like this allows you to quickly assess which bed times produce the most sleep by individual day. As bed time to bed increased (12 = midnight, 13= 1am, etc…), hours slept decreased. With this client, she woke up at the same time, but the time to bed varied which affected sleep duration.
