Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

FAR EASTERN UNIVERSITY – DEPARTMENT OF MEDICAL TECHNOLOGY

DESCRIPTIVE STATISTICS
• A

I. Measure of Central Tendency


• A descriptive statistic that describes the average, or typical value of a set of scores
• Common measures of central tendency: mean, median, and mode x 𝐱̅ - x
92 -2
A. Mean 88 2
• The arithmetic average of all the scores (x)/N) 90 0
• The number (𝐱̅) that makes (𝐱̅– x) equal to zero (0) 87 3
• The mean of a population is represented by ; the mean of a sample is by 𝐱̅ 94 -4
• Calculating the Mean (𝐱̅) 86 4
1. Sum the scores (x)
93 -3
90 0
2. Divide the sum (x) by the number of scores (N) = x/N
95 -5
• Used when the data are interval or ratio (scaled) that is not skewed 85 5
B. Median x = 900 (𝐱̅– x) = 0
• The 50th percentile, 2nd quartile or middle score x/N = 90
• Half of the scores are larger (and smaller) than the median
• Calculating the Median
1. Sort the data from highest to lowest
2. Find the score in the middle
a. Score in the middle = (N + 1) / 2
b. If N is even, the median is the average of 2 scores in the middle
What is the median of the scores? What is the median of the scores?
10 8 14 15 7 3 3 8 12 10 9 24 18 19 42 16 12
1. Sort the scores: 1. Sort the scores:
15 14 12 10 10 9 8 8 7 3 3 42 24 19 18 16 12
2. Find the score in the middle: 2. Find the score in the middle:
Score in the middle = (11 + 1) / 2 = 6th Score in the middle = (6 + 1) / 2 = 3.5
• Median = 9 • Median = (19 + 18) / 2 = 18.5
• Used when the distribution of scores is either positively (few really large
scores) or negatively (really small scores) skewed
C. Mode
• The score that occurs most frequently in a set of data
• Other distributions
1. Bimodal - distribution has 2 modes
2. Multimodal - distribution has >2 modes
• Insensitive to changes in the data set as 2 data sets can have the same mode
• Primarily used with nominal and ordinal data
D. Relations Between the Measures of Central Tendency
• For normal distributions, the mean = median = mode
1. positive (right) skew - the value of the mean is greater than the median
2. symmetrical distribution - the value of mean and median are equal
3. negative (left) skew - the value of the mean is smaller than the median

• A

AARON JAN PALMARES, RMT, MSMST_09.11.19 Page 1


FAR EASTERN UNIVERSITY – DEPARTMENT OF MEDICAL TECHNOLOGY

II. Measure of Dispersion


• Descriptive statistics that describe how similar a set of scores are to each other
• The more similar the scores, the lower will be the measure of dispersion
• The less spread out a distribution, the lower the measure of dispersion
• 3 main measures of dispersion: range, semi-interquartile range (SIR), and variance/SD
A. Range
• The difference between the largest (XL) and the smallest score (Xs) in a date set
• Range = XL - XS
What is the range of the data?
4 8 1 6 6 2 9 3 6 9
1. Range = XL - XS = 9 - 1 = 8
• Used when the data is ordinal or if your presenting results to people with no knowledge of statistics
• Insensitive as 2 very different sets of data can have the same range: e.g. 1 1 1 1 9 vs 1 3 5 7 9

B. Semi-Interquartile Range
st th rd th
• The difference of the 1 (25 ) and 3 quartiles (75 percentile), divided by two
• Calculating the SIR
1. Sort the scores from lowest to highest
2. Find and calculate the Q1 [= ¼ (N+1)] and Q3 [= ¾ (N+1)]
3. Calculate the SIR = (Q3 - Q1) / 2
What is the SIR of the data?
22 12 14 7 18 16 11 15 12
1. Sort the scores:
7 11 12 13 14 15 16 18 22
2. Find and calculate the Q1 and Q3:
• Q1 = ¼ (9+1) = 2.5
Q3 = ¾ (9+1) = 7.5
nd
• Q1 = 2 score is 11 + [0.5*(12-11)] = 11.5
th
• Q3 = 7 score is 16 + [0.5*(18-16)] = 17
3. SIR = (Q3 – Q1) / 2 = (17 – 11.5) / 2 = 2.75
• The SIR is used with skewed data as it is insensitive to the extreme scores

C. Variance
• Variance is the average (mean) of the squared deviation score
1. Population variance (σ2 - sigma)
2. Sample variance (s2)
• The average of the squared differences from the mean.
• Calculating the variance
1. Subtract the mean from each of the scores
x x - 𝐱̅ (x - 𝐱̅)2 Computations
92 -2 4 N = 10
2. The difference is called a
deviate or a deviation score 88 2 4 𝐱̅ = x/N = 90
• A measure of dispersion for a given score 90 0 0 (x - 𝐱̅)2 = 108
(How far away is a score from the mean?) 87 3 9
-4
σ2 = (x - µ)2 /N = 10.8
3. Square the deviation score (to make all scores 94 16
positive), then get the average (x - 𝐱̅) /N
2 86 4 16
-3
S2 = (x - 𝐱̅)2 /N-1 = 12
• The larger the variance, the more the scores 93 9
deviate from the mean 90 0 0
95 -5 25
85 5 25

AARON JAN PALMARES, RMT, MSMST_09.11.19 Page 2


FAR EASTERN UNIVERSITY – DEPARTMENT OF MEDICAL TECHNOLOGY

D. Standard Deviation
• Represents the “average” distance from the
mean or variability of a population
1. Standard deviation = variance Standard Deviation and Coefficient of Variation
2. Variance = standard deviation
2
 = 3.29 𝑺 = 3.46
• Expressed in the same units as the data. CV (%) = 100/µ= 3.66% CV (%) = 100S/x̅ = 3.84%
• Used to measure confidence in statistical
conclusions.
• Calculating the Standard deviation (SD)
1.  (sigma) is the population SD,
 (mu) is the population mean
2. s is the sample SD
𝐱̅ is the sample mean

E. Measure of Skew
• Measure of asymmetry in the distribution of scores
• Formula for determining the skew (See Figure)
1. If s3 < 0 (negative skew)
2. If s3 > 0 (positive skew)
3. If s3 = 0 (symmetrical)
4. The more different s3 is from 0, the greater the
skew in the distribution

F. Kurtosis
• The sharpness of the peak of a frequency-
distribution curve.
• Measures whether the scores are spread out more
or less than in a normal (Gaussian) distribution
i. Leptokurtic - When the distribution is less
spread out than normal (kurtosis is >3)
ii. Mesokurtic - When the distribution is
normally distributed (kurtosis = 3)
iii. Platykurtic - When the distribution is more
spread out than normal, (kurtosis <3)

G. s2, s3, & s4


2 3
• Collectively, the variance (s ), skew (s ), and
4
kurtosis (s ) describe the shape of the distribution

Homework
1. Case 1: MT3012 has 10 students and their weight (in Kg) are
as follows: 50, 70, 75, 75, 62, 60, 62, 55, 65, 52. What is the
mean, median and mode?
2. Case 2. A management of a pharmaceutical company is in
discussion to start a new reward structure, but they first
want to know how much their production spread is. The
management has collected their average daily production
data for last 10 days per (average) employee. 155, 169, 188,
150, 177, 145, 140, 190, 175, 156. Compute for the S.I.R.

AARON JAN PALMARES, RMT, MSMST_09.11.19 Page 3

You might also like