In 2007 Microsoft Word changed from using Times New Roman as the default font to Calibri as the default font. Microsoft Office 2010 and 2013 continue to use Calibri. When an expository writing class specifies writing a two page or ten page paper, the change in the font metrics can change the number of words that have to be written. One way to measure this is to compare the number of characters per line for each font using the same text.
The larger the number of characters per line, the more words one must write to fill the required number of pages. The smaller the number of characters per line, the fewer words one must write to fill up the required number of pages. The number of characters for eight full lines of text were counted in Times New Roman and Calibri. The text sample can be seen below.
Times New Roman is a serif typeface commissioned by the British newspaper The Times in 1931, created by Victor Lardent at the English branch of Monotype. Calibri is a humanist sans-serif typeface family under the Microsoft ClearType Font Collection, designed by Lucas de Groot for Microsoft. - Wikipedia, accessed on 24 March 2013.
| Times New Roman | Calibri |
|---|---|
| 91 | 91 |
| 93 | 86 |
| 100 | 98 |
| 103 | 100 |
| 90 | 93 |
| 88 | 88 |
| 96 | 96 |
| 102 | 97 |
| Question | Times New Roman | Calibri |
|---|---|---|
| 1. Calculate the sample size n for each sample. | ||
| 2. Calculate the sample mean x for each sample. | ||
| 3. Calculate the sample standard deviation sx for each sample. | ||
| 4. Calculate the standard error of the mean (SE). | ||
| 5. For a confidence level of 95%, calculate the t-critical. | ||
| 6. Calculate the margin of error E for the mean. | ||
| 7. For a confidence level of 95%, calculate the lower bound for the mean for each sample. | ||
| 8. For a confidence level of 95%, calculate the upper bound for the mean for each sample. | ||
| 9. Are the mean number of characters per line mathematically different for Times New Roman and Calibri? | ||
| 10. Based on your analysis above, is the difference in the means statistically significant at a confidence level of 95%? | ||