You are here
Group Members: Maria, Victoria, Trevor
Analyze the correlation between hours slept per week and GPA.
Make a scatterplot with hours slept per week on the x axis and GPA on the y axis to visualize the relationship.
Note any outliers
Our group (#2) chose to focus on the hours slept and hours studied in relation to GPA. We determined that making 2 graphs showing the gradual change in GPA with both hours slept as well as hours studied would change the average GPA per student, regardless of gender. The independent variables in these graphs would be either the hours slept or the hours studied. Upon analyzing the data we suspect that the hours spent studying, nor the hours one spent sleeping are significntly correlated to the GPA of the students. We could do this by comparing the overall graphs or the averages of 2 distinct groups such as the top 10 and the bottom 10 students. In addition, we could also create a normal distribution of the data and establish a standard deviation from the class average. From there, we could determine who are outliers and then assess their hours slept and studied to see if there are any obvious differences that may have led to this variability.
This data set has four columns: gender, GPA, hour studies, and hours slept per week. Just skimming the chart, we tried to look for correlations between hours studied and GPA and hours slept and GPA. This is difficult to do in this format and would be much easier to analyze with a graph, as we are studying more than one variable. In order to properly study these variable, assuming that the hypothesis of the study was that more sleep correlates to a higher GPA and more hours studied correlates with a higher GPA, we would make two different scatterplot graphs comparing these two sets to each other. In both cases, GPA would be the dependent variable. In the end, we made scatterplot graphs and found that there was a positive correlation between hours studied and GPA, however there was no correlation between hours slept and GPA. We have been able to identify a couple of outliers that do not support this hypothesis, as the individual with the most hours studied (50) has a GPA of 1.12.
Matthew Parkllan, Liam Gorman, Austin Meserole
One way that these data can be interpreted is by creating two different graphs, one with GPA and Hours studied as the x and y axis, and have another graph with hours studied vs hours slept as the x and y axis. If we wanted to be really thourough we could make these two graphs each for both male and female students. Sorting the data out this way would be a good way to prove or disprove the hypothesis that more hours of sleep leads to a higher GPA.
The data presented to us was from Karakuri. This data illustrates individuals gender, GPA, Hours studied, and Hours slept per week. To find the correlation between two things like gender vs GPA it could be depicted in a bar graph. Another type of correlation would be between hours studied versus GPA's. After analyzing a portion of the data presented it could be seen that males or females who studied more had a higher GPA, however, there were outliers present. In addition, these individuals also had a lower number of hours slept.
We would make a scatter plot of hours studied vs GPA. On the plot we would include a line of regression for the data of only males, one for the data of only females, and one for everybody overall. This would show the relationship between the hours studied and the GPA, as well as showing if there were any large differences in the relationship between males and females. We would also make a scatterplot of hours of sleep vs GPA, to get an idea of that relationship.
A research project was conducted on several different islands resulting in the data below:
• Install R and Rcmdr. Refer to R Commander Installation Notes for details: http://socserv.socsci.mcmaster.ca/jfox/Misc/Rcmdr/installation-notes.html • Import the data: ◦ In Rcmdr. under “Data” menu “Import data” from “text file” ◦ Set the “Field Separator” to “Commas”. ◦ Navigate to the CSV file and select it. • Click the “Edit data set” button to open the data set in a window.. ◦ Make a note of all outliers (to put in the legend of the figure). ◦ Click on the number of each row with an outlier, then right-click and “Delete current row”. ◦ Click OK to save edited data set. • Under “Graphs” choose “Scatterplot matrix...” ◦ Select all three variables. ◦ Click “Plot by groups”, select Gender, and click OK. ◦ Click “Options” and select the checkbox for Least-squares line and click OK.
Their are a few similarities and differences between these two figures on page 14. Both figures include two birds eye view pictures and one on the ground of roughly the same location, but the exact locations, zoom levels, and times of day appear to be the most obvious differences. The first figure's ground level picture is on a cloudy day and you can see the church, the library, and 3 fountains of the campus pond while the second figure's ground level picture is on a bright sunny day and only includes two fountains, the church and the library. Out of the two birds eye pictures of each figure, one shows a more developed metawampe lawn with grass and concrete ramps while the other picture of each figure has what seems to be just pathes of dirt in certain areas. Another similarity between the two birds eye pictures that i noticed is that th foliage around the campus pond between the two figures almost make the campus pond look like a different shape, like trees were added and they hide the real waterfront of the campus pond. Another difference between the two birds eye shots in each figure is that the one with more dirt patches also seems to include small buildings that are replaced with grass in the other shot.
One inference that I think can be made is that the two birds eye shots of each figure are to compare the before and after. Today the metawampe lawn and the other grass areas looks like they do right now while the other resembles what I would guess the areas looked like before development.
In panels A, B and C the lighting is different between the original and replicate. The original panels from are darker in color with more saturated tones. The foliage in the original panels and the replicates appears different. The replicates show the plants with fuller foliage in a lighter shade of green. The angles are also different. In each panel the original is from a lower angle and does not include the entirety of the pot which the replicates do; these original photographs are more level with the edge of the pot. The letters on the replicate panels have a period after them, while the originals do not.
The different angles of the photographs also vary what is included on the sides of the pictures. The original panel A shows a space between the bamboo backing so the outside is visible. The replicate does not have this space, but shows the ground and the bottom of the bamboo backing; it also shows a red rectangular object behind the plant and water on the table.
The original panel B does not show any of the table; it is the plant, the top of the plant, a portion of a plant next to it and the bamboo backing. The replicate shows the same things, but additionally shows the entire pot, the table and a red rectangular object behind the table.
The original panel C shows the top of the pot up with a very small portion of the plant next to it included. The replicate shows the entire pot, the table and much more of the plant next to it and a plant on the other side.
The light differences could be based in different times of day/season or different equipment. The angles and format were probably not specified in the methods.