Lecture Note Information
Purpose of the Course:
The purpose of this course is to provide an introduction to research methodology in the
social sciences. This class will cover both qualitative and quantitative approaches to
Quantitative Vs. Qualitative:
Quantitative – applied statistics, numbers
Qualitative – interviews, research, everything else
Experiments Vs. Surveys:
Experiments – very controlled environment – Pretest, Treatment, Post-test
Survey – ask them a number of q‟s – do not manipulate subjects – each question on
survey represents a variable
Types of Variables:
Quantitative Variables – continuous (rank), numerical
(e.g. education… #of years of schooling and income…# money)
Categorical – non-rank order, non-numerical
(e.g. religion… Catholics do not rank higher than protestants and race…skin colour)
When the mean is equal to the median (the average is equal to the middle), this is a
Normal distributions allow us to conduct significance tests.
The distribution of a variable using data from our sample is called sample distribution.
The distribution of the variable in the population is called population distribution.
The distribution of the mean of a variable from all possible samples is the sampling
Sampling distribution – as its standard deviation gets smaller as the size of the sample get
Sampling distribution – based on calculating a statistic over and over again from sample
samples of the same size.
Measures of Central Tendency:
Mean, Mode, Median, Range
Refresher: mean = average, mode = most occurring, median = middle (when written out
from least to greatest), and range = largest number subtracted by smallest number
Variance/Standard Deviation: (N = so in other words, the sample size)
Standard deviation is the square root of the variance.
To calculate the standard deviation:
1) calculate the mean
2) for every observation:
a. first observation minus mean = value. Square that value = new squared
b. Second observation minus mean = value. Square that value = new squared
e. last observation minus mean = value. Square that value = new squared
3) Add all the squared values to new a new value (sum of squared deviations).
4) Divide that value by the sample size (n) minus 1.
Validity and Reliability:
(Kayley – go to Brad‟s notes for this – UoG notebook)
The difference between the statistic and the parameter due to random process is known as
Random samples have less error than nonprobability samples. The sampling error
decreases as sample size goes up.
Central Limit Theory:
- regardless of the population distribution, with repeated sampling, the shape of the
sampling distribution is approximately „normal‟ (or bell shaped) with the
population parameter at its centre.
Law of Large Numbers:
- the larger the samples, the more the confidence because it‟s closer to the
- z scores = take „x‟, subtract the mean and divide by the standard deviation –!Tis a specified valuece between
–!Test for the difference between
•!Testtwo valuesd on the sampling
•!Tests are based on the samplingmany standard deviations an observation is from the mean
•!z-score-u* = p