A result that did not match the prediction

Text A: an original contemporary practice extract.

Our class predicted that the shaded puddle would last longer than the one in sunlight. By lunchtime, the shaded one had almost gone. We initially thought we had recorded the positions incorrectly. Then someone noticed that its water had been draining through a crack beside the kerb. The two puddles had differed in more than the amount of sunlight.

We wrote down the first result rather than erasing it, then chose two shallow containers with the same shape and amount of water. Their positions could be checked at set times. The second observation did not turn our first mistake into a successful prediction. It gave us a way to ask a more specific question after discovering what the first comparison had left uncontrolled.

Text B: an original contemporary practice extract.

The volunteer trial compared how clearly two signs could be read from the back of a hall. The team expected the larger lettering to be easier, but more people misread that sign. Instead of declaring size unimportant, the organiser inspected the photographs of the test. The larger sign had been placed beside a bright window, while the smaller one hung against a plain wall.

Both signs were tested again on the same wall with the same lighting. The report included the earlier results and explained why they could not settle the question of letter size alone. The surprise had been useful because it exposed a difference in the conditions, not because it proved that the original prediction must be wrong in every setting.

Both texts describe the investigations into unexpected results.

What similarities do the investigations into unexpected results share in these extracts? Use evidence from both texts to support your answer.