Also the results of this research cannot be generalised. The findings of this study can be only applied to students at this age (13-15 years old), and within the same level of education. As the students between the top and bottom sets were not tested, the results may not be generalised to the same age group. In order to over come this problem a bigger sample size may be used using a wider age range of participants from different abilities.
Although any participants that were not born in the UK were excluded from the experiment, other participants came from very different backgrounds, which may act as a confounding variable as their first language might be different, as well as their background study and interests, as those may be influenced by culture. In order to over come this problem, the participants may be asked before the experiment what language they speak at home and whether they have been studying only inside the UK since year 1 for example, so that all the participants have the same level of education.
Another confounding variable could be the randomisation process, as it was not appropriately applied as the males and females were hap hazardously chosen from each set. If I were to repeat this experiment I would divide the students into two groups (males and females) and allocate a number to each participant and then choose their number randomly using a computer in order to give each participant an equal opportunity of being chosen.
Demand characteristics may had also taken place, as four participants were taken from the same class and the participants could not have been tested at the same time, therefore some participants could have been informally told by others about the tasks, which might had led them to be mentally prepared for the task and hence make them be at a more advantageous position than others. However the participants" names were not announced until they were taken to perform the test.