Skip to content
This repository has been archived by the owner on Jun 17, 2022. It is now read-only.

Add AB test metrics eda #88

Closed
wants to merge 46 commits into from
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
46 commits
Select commit Hold shift + click to select a range
7c5d3af
Add AB test metrics eda
mammykins Jan 28, 2019
f94d190
Fix convenience
mammykins Jan 28, 2019
30254d5
Fix failed copy and paste
mammykins Jan 28, 2019
3a665e4
adding more metrics, and space for ones I haven't finalised extractio…
Jan 28, 2019
6b116fd
some notes on nav events
Jan 28, 2019
0a62291
don't print so many example events
Jan 28, 2019
8b0b1c6
some plots about how many related clicks journeys tend to have
Jan 29, 2019
7b5aeb9
more notes about nav events
Jan 29, 2019
48b5cca
exploring event categories
Jan 29, 2019
2926586
nav events notes
Jan 29, 2019
912db92
rearranging some cells and adding notes
Jan 29, 2019
cfd85a2
rerunning some things
Jan 31, 2019
2377500
some work sampling a more representative file (no dropping one off jo…
Feb 1, 2019
e7eff41
looking at page events finding vs thing
Feb 1, 2019
b30aa3b
doing a query of document type, so we can map it to pages
Feb 1, 2019
f96904f
neatening up WIP generating data notebook
Feb 4, 2019
4c41a51
working on sampling
Feb 4, 2019
2aa8ba7
get rid of long Counter
Feb 4, 2019
b09be45
more sampling investigation
Feb 4, 2019
c4f90e0
add finding things query to general notebook
Feb 4, 2019
57e3f4e
looking at event page tuples
Feb 4, 2019
1bc1324
looking at nav events and searching after content pages
Feb 5, 2019
cd302a4
notes about speeding up search analysis
Feb 5, 2019
48bc2ef
search pages after content pages
Feb 6, 2019
f8550ba
explain search pages work a bit
Feb 6, 2019
a57a9f5
explain search pages work a bit
Feb 6, 2019
2fc41af
can't find a better search after content detecting fn with this data
Feb 8, 2019
3fb6dc7
one (slightly) less cluttered notebook
Feb 11, 2019
adaa74a
Use derived metrics for z_prop test
mammykins Feb 12, 2019
faa1176
Add bayesian bootstrap workflow
mammykins Feb 12, 2019
59f654b
Identified bug to fix
mammykins Feb 13, 2019
8a0415b
Complete metric derivation and analysis
mammykins Feb 14, 2019
f4fe5a1
Add comment describing metadata dir
mammykins Feb 14, 2019
06b0095
adding more metrics, and space for ones I haven't finalised extractio…
Jan 28, 2019
d73053d
some plots about how many related clicks journeys tend to have
Jan 29, 2019
7e951b9
Add bayesian bootstrap workflow
mammykins Feb 12, 2019
9d2431a
Identified bug to fix
mammykins Feb 13, 2019
3a67434
Complete metric derivation and analysis
mammykins Feb 14, 2019
735c3f4
Add comment describing metadata dir
mammykins Feb 14, 2019
c63cf14
update README about analysis
Feb 15, 2019
e2f842b
exploring event categories
Jan 29, 2019
1a05d7a
some work sampling a more representative file (no dropping one off jo…
Feb 1, 2019
94cc764
looking at page events finding vs thing
Feb 1, 2019
6697b6c
looking at nav events and searching after content pages
Feb 5, 2019
336937b
Use derived metrics for z_prop test
mammykins Feb 12, 2019
cee1f90
Merge branch 'squashed_analysis' into ab_test_metrics_eda
Feb 15, 2019
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -188,6 +188,13 @@ Here's some definitions of the columns in the resulting dataframe:

For help getting started analysing and visualising this type of data using R or Python,
see our notebooks in `notebooks/eda`.

## Using processed journey data for A/B tests
See the [govuk_ab_analysis](https://github.com/ukgovdatascience/govuk_ab_analysis) repo for some scripts to do this analysis.

`notebooks/eda/generate_ab_rl_mvp.ipynb` is a notebook to analyse an A/B test, using data generated by the `stnd_taxon_ab.sql` query.
`notebooks/eda/generating_ab_test_data_with_workings.ipynb` is a notebook with some more workings of how we derived our metrics.


# Converting processed_journey data to functional network data

Expand Down
Loading