# Presentation2

Views:

Category: Education

## Presentation Description

No description available.

## Presentation Transcript

### Slide 1:

This dataset represents the grades achieved by 74 different students on level 1 modules last year.

### Slide 2:

The data can be represented in a chart. The x-axis (horizontal) indicates which student the grade belongs to, and the y-axis (vertical) indicates the grade itself. You can see here that the 36th student, achieved a grade of 51%

### Slide 3:

Once we have our main dataset, we can always map it against other data. For example, here I have taken the attendance % for each student, and matched that figure against their grade. For example, the student who recorded a grade of 54%, also recorded attendance of 90%. The student who recorded a grade of 93%, had a 100% attendance record.

### Slide 4:

We can now map both sets of data on the same chart. You may notice, each dataset has a clear upward trend, which suggests the two sets of data correlate with each other. In other words, the higher the grade, the higher the attendance…

…and visa versa

### Slide 6:

Let’s try another set of data. Here, we have paired the grades of each student with the number of sources listed in the bibliographies of their essays. So, for example, the student with a grade of 68% listed 5 sources in their bibliography.

### Slide 7:

Highest value = 11 5 / 11 X 100 = 45% 11 / 11 X 100 = 100% When we represent this data on the same chart, we have a problem. The highest value for the ‘sources’ data is 11. The highest value for grades is 93. The sources data is squashed down at the bottom and it is hard to see.

### Slide 8:

Highest value = 11 5 / 11 X 100 = 45% 11 / 11 X 100 = 100% What we need to do, is to convert the ‘sources’ data in to a % value. This means determining each value as a % of the highest value (11). If 11 is 100%, then 5 is 45%*. *Actually, it would be better to calculate 11 as 93%, but this way just makes it easier for now.

### Slide 9:

Now the data has been converted, we can map it onto the same chart as the grade data. Highest value = 11 5 / 11 X 100 = 45% 11 / 11 X 100 = 100% As you can see, not only is the data now much clearer, we can see again the ‘upward trend’ which suggests a correlation. Conclusion? The more sources you use, the higher your grade!

### Slide 10:

57 / 57 X 100 = 100% 28 / 57 X 100 = 49% So what about age? Is this a factor in whether you get a good grade or not? Well, here is the age data which has again been mapped against the grades. As with the sources data, the age data needs to be converted to a % value in order to match the grade scale.

### Slide 11:

These are the converted values, which can now be mapped onto the same chart again. Crumbs. The wildness of this squiggly line gives the data a rather random effect. We can see this more clearly if we compare the trends of each set of data… The trends are clearly not going in the same direction. Conclusion? There is no correlation between the age of the student, and their grade.

### Slide 12:

So what have we found out? Well, we have seen that there are two things which definitely seem to be closely related to whether you get a good grade or not: Attendance, and reading / using a wide range of sources.

### Slide 13:

However, age does not appear to be a factor. Younger students are no more likely to get a higher grade than older students, and older students no more likely to get a higher grade than younger students.

### Slide 14:

There are some problems with data of this sort. For example: How do I know last years students were not just a bunch of weirdoes? How do I know NUC is not just a weird environment? How do I know that the it is not simply that the higher the grade, the more able the student, and that more able students tend to turn up to more classes? How reliable is the data itself? The grades can be verified by checking the files, but not all assessments are essays, which contain bibliographies – and class registers can contain errors. How far have I gone to ensure the reliability of this data?

### Slide 15:

Conclusion? What next?