Leave a comment, click here
MY
INTRODUCTION
MY
HYPOTHESIS
MY NULL
HYPOTHESIS
MY
QUESTION
MY
THEORY
MY
RESEARCH
MY
CONCLUSION
From our initial data,we are able
to summarize a few points.
My
results
Mean & Median
Skewness
Dispersion
Mode
Quantiles
Back to my research
The Mean is the sum of all the values, divided by how many values there are. I If you have 3, 3, 4, 5, and 1, mean = 3+3+4+5+1 5 = 3:2.
The median is the value exactly in the middle. Imagine it as the balancing point.
Our median is to the right of our mean, indicating more agreement with a shift in polarization.
Professor Jane Lawrence-Sumner, POL 3085, Quantitative Analysis in Political Science, February 18, 2020
Mode is the most common (or most frequently occurring value). Very useful for categorical or dichotomous data. Useful for ordinal. Not very useful for continuous, since it would treat, as an example: $53,000 and $53,0001 as different.
As you can see, our mode leans toward
4 (Moderately) or 5 (Very much)
Professor Jane Lawrence-Sumner, POL 3085, Quantitative Analysis in Political Science, February 18, 2020
Skewness
Skew: the distribution is not symmetrical. In other words, there are outliers.
Negative skew: the outliers are mostly below the central tendency.
Positive skew: the outliers are mostly above the central tendency.
Our skews are negative, indicating a longer tail to the left, as shown in the yellow bell curve.
Professor Jane Lawrence-Sumner, POL 3085, Quantitative Analysis in Political Science, February 18, 2020
Dispersion is how spread out are data?
In addition to the central tendency (what is the typical value), we often want to know how spread out the data are.
Many measures of this: most common are standard deviation and variance, the one I used.
What matters: very spread out data (high standard deviation, high variance) can be very different from narrowly spread out data.
The variance here are between 1.17 and 1.35 which appears low, however this is relative to the scale, in this case 1-5, 23.4 percent - 27.0 percent.
Professor Jane Lawrence-Sumner, POL 3085, Quantitative Analysis in Political Science, February 18, 2020
Quantiles are another way to figure out how spread out your data are by showing you how many quarters of the data is lower than a certain value.
Our quantiles match the graph.
Professor Jane Lawrence-Sumner, POL 3085, Quantitative Analysis in Political Science, February 18, 2020