Scribd
Upload
11 views32 pages

PSM 2k23

The document outlines key statistical concepts including the Student t-test, Chi-square test, probability distributions, and measures of central tendency and dispersion. It provides formulas for calculating confidence intervals and describes types of variables and their measurements. Additionally, it discusses the properties of normal distribution and skewness in data.

Uploaded by

opaleyeismail9
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
11 views32 pages

PSM 2k23

The document outlines key statistical concepts including the Student t-test, Chi-square test, probability distributions, and measures of central tendency and dispersion. It provides formulas for calculating confidence intervals and describes types of variables and their measurements. Additionally, it discusses the properties of normal distribution and skewness in data.

Uploaded by

opaleyeismail9
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 32

PSM

2k23 REVISION
PART 2

-Jegede Olaoye Ajiboye[Jegz]


HIGH YIELD TOPICS
• 1. Student t-test and confidence interval
• 2. Chi-square
• 3. Normal distribution Curve
• 4. Measures of dispersion/variation
• 5. Measure of Central tendency
• 6. Probability distribution
• 7. Binomial distribution
• 8. Proportion
• 2013 Q2b
• Steps in Testing of Statistical Hypothesis: Student t-test and Chi-square test
1. State the null hypothesis
2. State the alternative hypothesis
3. State the level of error
4. Choose the appropriate test statistic
5. Calculate the degree of freedom: t-test df = n 1+n2-2, Chi-square df= (r – 1) (c – 1)
6. Evaluate the test statistic
7. Decision making and Conclusion
Student t test - types
• One sample t test
• Comparing the mean of a sample with a previously known mean
• Independent samples t test
• Comparing the mean of two samples that are unrelated or independent
• Paired t test
• Comparing the means of related samples
• Data situations involve matched study designs
• Other data situation is when a set of twins are divided into two groups (with one twin
in each group)
• A third data situationBefore and after designs (pre and post designs)
Student t test - types
• One sample t test
• Degree of freedom is (sample size – 1)
• Independent samples t test
• Degree freedom is total sample size - 2
• df = n1 + n2 – 2
• Paired t test
• Degree freedom is number of pairs - 1
Calculation of the 95% CI for
difference in means
• 95% CI for the difference in means
• ( x1 – x2) +/- 1.96SE (x1 – x2)

• (x1 – x2) +/- 1.96√(S12/n1 +S22/n2)


Chi-square
Χ2 = Σ[(Oi-Ei)2/Ei]

where:
Oi = Observed frequencies
Ei = Expected frequencies if the null hypothesis were true
d.f. = (r-1)x (c-1)
Probability distribution

• Can be discrete or continuous probability distributions based on the nature of the random
variable
2012 Q1b, 2011 Q2a, 2013 Q1a

• Discrete probability distributions – for discrete random variables (count data)


• Discrete probability distribution is a type of probability distribution that shows all possible values of a
discrete random variable along with the associated probabilities. In other words, a discrete probability
distribution gives the likelihood of occurrence of each possible value of a discrete random variable.
•Examples
– Binomial distribution
– Poisson distribution
– Geometric distribution
•Others: Hypergeometric distribution, Multinomial Distribution, Negative binomial distribution.
• Continuous probability distributions – for continuous random variables
– Normal distribution
– Exponential distribution
– Logistic distribution
Probability: Binomial distribution

• The probability of ‘r’ successes out of ‘n’ trials of a binomial experiment is given by the formula
• Pr (r out of n) = nCr pr qn-r
• = n! pr qn-r
• (n-r)! r!
• n is the total number of trials or attempts
• r is the number of successes
• P is the probability of success
• q is the probability of failure
•2013 Q1b, 2014 Q2c
•Parameters of Binomial distribution are
i. the number of trials (n)
ii. the probability of success (p)
•
•2013 Q1c
i. mean is np
ii. variance np(1- p)
• Standard deviation= npq
•2013 Q1d
•The binomial distribution is appropriate here because only two outcomes are possible: died or
survived
• p = 36% or 0.36 hence q = 1- 0.36= 0.64
• n = 6 cases
• r for the first question is 0, it’s 3 for the second and takes more than one value for the third
question (4 or more = 4, 5 and 6)
•Pr (r out of n) = nCr pr qn-r
•= n! pr qn-r
r! (n-r)!
a) Pr (0 out of 6 cases) = 6! 0.360 x 0.64(6-0)
0! X 6!
=
b) Pr (3 out of 6 cases) = 6! 0.363 x 0.64(6-3)
3! X 3!

c) Pr (4 or more) = Pr (4) + Pr (5) + Pr (6) OR 1- Pr (0) + Pr (1) + Pr (2) + Pr(3)


LEVEL/SCALES OF MEASUREMENT

•Nominal, Ordinal,Interval,Ratio
VARIABLES

•Types of variables
• Quantitative variables (numeric data)
– Have numerical dimension
• Qualitative variables (attributive or categorical variables)

• No numerical dimension
2012 Q2
QUALITATIVE VARIABLES

– Can be classified based on number of categories of variable as


• dichotomous (2 categories) e.g. gender, Exam result(pass/fail),
outcome of coin tossing( head or tail)
• multichotomous (more than 2 categories) e.g. blood group,
genotype,race
– Could be ordered (severity of injury, stage of cancer) or unordered
(ethnicity)
QUANTITATIVE VARIABLE

• Examples: Weight, height, radiation levels, humidity, pain scores, temperature


– Can be organized into:
i. Discrete (only whole number values possible) e.g., Age(last birthday), red blood cell
counts, bacteria count, Apgar’s score
ii. continuous (can take decimal values) e.g., blood pressure, height in meters, blood
glucose, weight
2013 Q1a

•The normal distribution is used when the population distribution of data is assumed
normal. It is characterized by the mean and the standard deviation of the data.
•A sample of the population is used to estimate the mean and standard deviation.
•The t statistic is an estimate of the standard error of the mean of the population or
how well known is the mean based on the sample size.
•Which to use depends entirely on the question you are trying to answer. If the
question concerns the entire population as it is distributed, then the normal distribution
should be used. If the question concerns the mean of the population then the t statistic
may be used. For the use of either, a larger sample size gives a better result.
•
2014 Q2a

•Properties of the normal curve


• It’s bell-shaped
• It’s symmetrical about the mean value
• Determined by the mean and variance
• It’s mean, median and mode are equal
• Total area under the curve is 1 (100%)
• 68% of total observation lie within 1SD (left and right) of the mean value
• 95% of total observation lie within 2SD (left and right) of the mean value
• More than 99% of total observation lie within 3SD (left and right) of the mean value
2014 Q2e(Similar to 2012 Q1c, )

•PCV less than 40


• Compute the z score corresponding to x = 40 using the formula
• Z = (x – μ)/ σ = (40-35)/5 = 1.0
• The area corresponding to Z = +1 is 0.3413
• Total shaded area is 0.5 + 0.3413 = 0.8413
PCV between 20 and 37

• Compute the z score corresponding to x = 20 and x = 37


• Check the corresponding areas to z = (20-35)/5 and z = (37-35)/5
• The z scores are -3.0 and +0.4 and the corresponding areas (from the mean to
the z value) are 0.4987 and 0.1554 respectively.
• The total area is 0.4987 + 0.1554 = 0.65
PCV greater than 38

Compute the z score corresponding to x = 38 using the formula


Z = (x – μ)/ σ = (38-35)/5 = 0.6
• The area corresponding to Z = +0.6 is 0.2257
• Total shaded greater than 38 will be 1-(0.5+0.2257)= 0.2743
2011 Q1

•Measures of central tendency


• Mean
– Arithmetic mean
– Harmonic mean
– Geometric mean
• Median
• Mode
2011 Q1e
Measures of dispersion

• Standard deviation
• Variance
• Range
• Mean absolute deviation
• Interquartile range
• Coefficient of variation
2013 Q2a

•Nominal Vs Ordinal
– Merely classifies or categorizes vs Has Rank or order property
– No notion of numerical magnitude vs Has nominal properties.
– Focus on attributes or qualities of the phenomenon vs Differences between any two levels cannot be quantified meaningfully. e.g. level of pain: Mild,
Moderate, Severe.
– Examples: Blood Group( A, B, AB,O), Sex(male, female),Socio-economic status, Marital Status, Site of Cancer,Type of Disease VS Stage of cancer disease,
Patient’s pain condition, Level of Education, Doctors’ Professional Status
2017 Q1a
2017 Q1a
SKEWNESS

•Skewed Distributions.
–Positive if mean is greater than median.
–Negative if mean is less than median.

•Mean-Median>0 implies positive skewness


•Mean-Median<0 implies negative skewness
Mean=Median implies symmetry
THANK YOU
• FOR CORRECTIONS AND SUGGESTIONS
• 08079025059
• [email protected]

You might also like