Class Notes
(807,932)

Canada
(492,930)

University of Toronto Scarborough
(30,795)

Psychology
(7,612)

PSYB07H3
(39)

Dwayne Pare
(22)

Lecture 3

# PSYB07 - Lecture 3 Notes.docx

Unlock Document

University of Toronto Scarborough

Psychology

PSYB07H3

Dwayne Pare

Fall

Description

PSYB07: Data Analysis in Psychology
Lecture 3 - September 26 , 2012
Descriptive Statistics
Describing Distributions
Modality – refers to the number of peaks in a data set
Mode: the most frequent term in the data set (the peak of the distribution)
Histogram – first indication of the data
Smoothing the data (drawing the line over bars) – gives a good indication of the
distribution’s shape
Classic bell curve/ the normal distribution
o Tail, body, tail
o Generally a peak in the center and symmetrical sides
o Proportionate clustering of the data
Unimodal – one peak or mode
o Common / normal
o Bell curve
Bimodal – two peaks or modes
o Heights of peaks could the same or different
Multimodal – more than two peaks or modes
o Appears messy
o To be avoided
o No pattern within the data
Amodal (uniform) distribution – no peaks or modes
o All values have the same frequency
Symmetry
An asymmetric distribution is skewed
o The body of the data is shifted to one side
o One of the tails is longer than the other
o Extreme or moderate skewness
o The direction of the skew is the direction that the tail is pointing
Positively skewed (towards right, or positive numbers)
Negatively skewed (towards left, or negative numbers) Kurtosis
Refers to data clustering
o How pointy or flat the shape of the distribution point
o Doesn’t affect measures central tendencies
But affects measures of spread or range
Mesokurtic – normal distribution
Leptokurtic – pointier than a normal distribution
o Less variability (in comparison to normal distribution)
Platykurtic – flatter than a normal distribution
o More variability (in comparison to normal distribution)
Descriptive statistics
The first way to describe the data
o Characteristics of samples and or population
2 main methods
o Measures of central tendency
The best measure/most representative to describe the data:
Mean – average of all scores
o The x value corresponding to the balance point of the
distribution
o Mean = sum of scores / number of scores
o “x bar” = sample mean and “mew” = population mean
o Not necessary to order the data to calculate mean
o Extreme values – affects statistics
o Large values in data will affect the mean
The mean is sensitive to extreme values
Known as outliers – which need to be removed
Median – the half point
o The midpoint of the distribution
o The x value corresponding to the midpoint
o Half of scores fall above, and other half fall below th
o 50 percentile
o Magnitude of values is not taken into account when
calculating the median
Is not affected by outliers/extreme scores – is only
concerned with that middle value
Meaning 50% of the scores fall below the median
(1) Arrange terms in numerical order
(2) Median location = (N+1)/2
(3) Median is the value at that location
Median does not have to be an actual score (in the situation of an even number of
scores)
(1) Order data
(2) Median location = (N+1)/2 = (the middle of the third and fourth term)
(3) Value of location 3.5 is the average of values at locations 3 and 4
Add values at locations 3 and 4 then find its average
That score is the median, even if it is not a score in the data
Mode – the point of highest frequency
o The most common score
o The x value corresponding to the peak of the distribution
o Also very sensitive to outliers/extreme scores
o In a normal distribution most data points are located near the middle of the
distribution
summarize your data with a single number representing the center
Stem/Leaf – no data points are lost / all terms are displayed and listed

More
Less
Related notes for PSYB07H3