Is the mean assets-to-liabilities ratio the same for healthy and failed companies?

This example describes a randomisation test.

Simulation
New data sets are generated directly from the model.
Randomisation
Modifications to the actual data are generated that would have been as likely as the original data.

The actual data are shown on the left. Randomisation uses the same values but randomly allocates them to the two groups. Click Randomise to randomly pick 33 of the the 101 values for the failed group.

The difference between the mean assets-to-liabilities ratios is the test statistic — it will be far from zero if H0 (no difference between healthy and failed companies) is false.

Click Accumulate and repeat the randomisation several more times. (Uncheck Animate to speed up the randomisation.)

A difference in means as far from zero as -0.902 is unlikely if the failed and healthy companies are really the same so there is strong evidence of a difference between the groups.

Emphasise the general concepts:

Null hypothesis, H0
Each assets-to-liabilities ratio would have been equally likely in a healthy or failed company
Alternative hypothesis, HA
Healthy and failed companies tend to have different assets-to-liabilities ratios
Test statistic
Difference between mean assets-to-liabilities ratios of healthy and failed companies
P-value
Probability of getting a difference that is so far from zero if H0 is true
Conclusion
A small p-value gives evidence that H0 is not true.

A study in Greece compared characteristics of 68 healthy companies with those of another 33 that had recently failed. The jittered dot plots on the left below show the ratio of current assets to current liabilities for each of the 101 companies.

The mean asset-to-liabilities ratio for the sample of failed companies is 0.902 lower than that for the healthy companies, but the distributions overlap. Might this difference be simply a result of randomness, or can we conclude that there is a difference in the underlying populations?