Baseline bar chart
Baseline charts are the heart of Protobi. Deceptively simple, they work the same for every survey data type, from categorical variables, grids, ratings, numeric open, text open ends, and more.
So named because they compare the current subset of respondents to a baseline population, highlighting statistically significant differences.
|Not too happy||15.4%|
|Don't know/Refused (VOL.)||3.2%|
Click to drill inClick on any value to select a subset of respondents.
All charts everywhere update to reflect this subset.
(option+Click on Mac)
The chart here shows respondents who replied "Very happy" are selected. The baseline distribution for reference is all respondents.
|Not too happy||0.0%|
|Don't know/Refused (VOL.)||0.0%|
Outlines show the baseline,
arrows show significant differences
This chart shows responses to q2 for those respondents above who selected "Very happy". Here we see respondents who are happy are significantly more likely to also be in "Excellent" health.
By default, all stats tests are at the 0.05 level of significance but you can change that.
|Don't know/Refused (VOL.)||0.3%|
Drag and drop crosstabs
Crosstab one element by dragging another "banner" element on to it.
Or cross all elements by dragging a "banner" element onto the Crosstabs button in the toolbar.
Choose between "Complement" testing (shown here) where each column is compared to all other
for simple highlighting
or "Pairwise" testing where each column is compared to each other column for detailed
Q.1 Generally, how would you say things are these days in your life - would you say that you are very happy, pretty happy, or not too happy?
|Not too happy||15.3%||6.0%||11.3%||29.6%||54.9%|
Bin numeric values into ranges
Numeric values are automatically binned into ranges for analysis, with the bin size based on the observed distribution and snapped to a nice multiple of 1, 5, 10 or 25.
You can choose your own bin ranges, including specific cut points.
This chart shows the BMI for US outpatient visits in 2016 by patients age over 15,
from US CDC National Ambulatory Health Care Survey
|11 to 15||0.1%|
|16 to 20||5.4%|
|21 to 25||23.9%|
|26 to 30||32.2%|
|31 to 35||21.3%|
|36 to 40||9.6%|
|41 to 45||4.6%|
|46 to 50||1.8%|
|51 to 55||0.8%|
|56 to 60||0.2%|
|61 to 65||0.1%|
Check-all-that apply questions are common in surveys but a little tricky to analyze since each option is an independent variable. Protobi automatically collapses these into concise summaries.
Compact to Yes
|Other (please specify)||9.4%|