As the experimental hypothesis predicted, the introverted participants’ average driving speed is lower than that of the extroverted participants’. As it can be seen from the chart, the confidence intervals partly overlap, which may suggest that the observed difference is due to the sampling error. However, one-tailed p-value is 0.037, which is less than 0.05 suggesting that the difference between the sample means is statistically significant (not due to the sampling error). The measure of effect size, which is the difference between two means expressed in terms of standard deviation, known as the statistic ‘d’, is 0.07. This can be considered a large effect, suggesting that the difference between the samples is due to actual differences between introverts and extroverts.
Descriptive statistics table
As it can be seen from the Chart 2, the Chart 4 and the descriptive statistics table, the mean number of emotions classified correctly by the UK’s participants is 10.70. Three UK participants classified all 12 emotions correctly. The median score is 10.50. By looking at the Chart 3 we can see that the most frequently (4 times) occurring value, the mode, is 10. The range of scores is 3 (12 – 9), with 9 being the lowest score of the UK sample. The standard deviation, the average deviation of each score from the mean score, is approximately 1. The 95 per cent confidence interval for the UK sample mean is 10.70 0.76, so we can be 95 per cent confident that the population mean value will fall between 9.94 and 11.46.
The mean number of emotions classified correctly by the Japanese participants is 9.2. The highest score, achieved by two Japanese participants, was 11. The median score is 9 and the mode is 8 (occurred 4 times). The range of scores is 3 (11 – 8), with 8 being the lowest score of this sample. The standard deviation is 1.2. The 95 per cent confidence interval for the Japanese sample mean is 9.20 0.88, so we can be 95 per cent confident that the population mean value will fall between 8.32 and 10.08.
The mean number of emotions classified correctly by the Chilean participants is 7.4. Only one participant correctly classified all 12 emotions. The median score is 8 and the mode is 8 and 4 (both occurring 2 times). The range of scores is 9 (12 – 3), with 3 being the lowest score of this sample. The standard deviation is 3.2. The 95 per cent confidence interval is 7.40 ï¿½ 2.29, so we can be 95 per cent confident that the population mean will fall between 5.11 and 9.69. Wide dispersion of Chilean scores is reflected in the values of the standard deviation and the confidence interval.
As it can be seen from the Chart 2, the Chart 4 and the descriptive statistics table, the UK sample has the highest mean, median and mode scores of correctly classified emotions. The UK sample has also the highest number of participants who classified all 12 emotions correctly. By contrast, the Chilean sample has the lowest mean, median and mode scores. The Chilean sample is characterised by a wide range of scores with one score of 12 but also one score of as low as 3 and couple of scores of 4. Scores of 3, 4 and 5 are the lowest scores of the study and occur only in the Chilean sample.
The Japanese sample, with scores slightly lower than those of the UK sample but generally higher and not as widely dispersed as those of Chilean sample, lays in the middle. The error bar chart for all the samples shows the overlap of the Japanese sample confidence interval with the Chilean sample confidence interval, from which we could conclude than any differences between them are due to the sampling error. This overlap is a consequence of large dispersion of the Chilean sample. The findings related to this sample could be explored and evaluated by a further study involving a larger sample.
Thomas, K (2002) The individual differences approach to personality, in Miell, D., Phoenix, A. and Thomas, K. (eds) Mapping Psychology, Milton Keynes, The Open University.