The document outlines the syllabus for a course on Statistical Data Analysis, covering topics such as parametric and non-parametric tests, descriptive statistics, t-tests, ANOVA, and chi-square tests. It includes definitions, examples, and applications of various statistical methods, along with sample problems and solutions. The content is structured to provide a comprehensive understanding of statistical analysis techniques used in research.
The document outlines the syllabus for a course on Statistical Data Analysis, covering topics such as parametric and non-parametric tests, descriptive statistics, t-tests, ANOVA, and chi-square tests. It includes definitions, examples, and applications of various statistical methods, along with sample problems and solutions. The content is structured to provide a comprehensive understanding of statistical analysis techniques used in research.
The document outlines the syllabus for a course on Statistical Data Analysis, covering topics such as parametric and non-parametric tests, descriptive statistics, t-tests, ANOVA, and chi-square tests. It includes definitions, examples, and applications of various statistical methods, along with sample problems and solutions. The content is structured to provide a comprehensive understanding of statistical analysis techniques used in research.
The document outlines the syllabus for a course on Statistical Data Analysis, covering topics such as parametric and non-parametric tests, descriptive statistics, t-tests, ANOVA, and chi-square tests. It includes definitions, examples, and applications of various statistical methods, along with sample problems and solutions. The content is structured to provide a comprehensive understanding of statistical analysis techniques used in research.
Assistant Professor Medi-Caps University, Indore Unit-IV Syllabus Statistical Data Analysis: Parametric and Non-Parametric Tests Descriptive Statistics Comparing Means using t-test - One Sample, Independent Samples and Paired Samples;
One-way Analysis of Variance (ANOVA)
Chi-square Test Parametric and Non-Parametric Tests
Parametric Test:- Parametric Test assumes that sample data comes from a population that has probability distribution with a fixed set of parameter. Example:- t-Test, Z-Test etc.
Non-Parametric Test:- The nonparametric test is defined as the
hypothesis test which is not based on underlying assumptions, i.e. it does not require population’s distribution to be denoted by specific parameters.
Example:- Chi- Square Test, Kruskal-Wallis Test
Difference between Parametric and Non-Parametric Tests
Properties Parametric Non-Parametric
Assumption Yes No Central Tendency Value Mean Median Value Correlation Pearson Spearman Probability Distribution Normal Arbitrary Population Knowledge Required Not Required Applicability Variable Variable & Attribute Example t-Test, Z-Test Chi-Square Test, Kruskal-Wallis Test Descriptive Statistics & Central Tendency
Descriptive statistics It is a means of describing features of a data set by generating summaries about data samples. It's often depicted as a summary of data shown that explains the contents of data. For example, a population census may include descriptive statistics regarding the ratio of men and women in a specific city.
Central tendency The central tendency is stated as the statistical measure that represents the single value of the entire distribution or a dataset. Tools of Descriptive Statistics
Mean :- The value which we get by dividing the total of the value of various given item in a series by the total number of item.
Median:- Median divides the data sets into two equal parts. Tools of Descriptive Statistics
Mode:- The mode is the value that appears most often in a set of data t-Test t-test A T-test is the final statistical measure for determining differences between two means that may or may not be related.
The t-test is based on the t- distribution, which
is a bell-shaped curve like the normal distribution, but has heavier tails.
As the sample size increases, the degrees
of freedom also increase, and the William Sealy Gosset becomes similar to the normal distribution. t-test Degree of Freedom Types of t- Test It is the number of values in the final calculation of a One Sample t- Test statistic that are free to vary. Independent Two Sample t- Formula of Degree of Test
Freedom = n-1 Paired (Two Sample) t- Test
Where, n = Number of Observation Welch’s Test or unequal Variance t- Test t-Distribution Curve Assumption in one Sample t-Test The Sample is Collected form Population through Simple Random Sampling method.
Each Observation (Sampling Unit) is independent to each
Other The Population Standard Deviation ‘σ’ is unknown.
The Population from which Sample is drown, follows
Normal Distribution Assumption in Two Sample t-Test Independence: The observations in one sample are independent of the observations in the other sample
Normality: Both samples are approximately normally
distributed.
Homogeneity of Variances: Both samples have
approximately the same variance.
Random Sampling: Both samples were obtained using a
random sampling method. Question Raju Restaurant near the Railway Station has average sale of 500 tea cup per day. Because of the development of bus stand nearby, it is expected to increase its sale. After the start of Bus Stand, the Daily sales of first 12 days is Following 550, 570, 490, 615, 505, 580, 570, 460, 600, 580, 530, 526 On the basis of above sample information, can one conclude that Raju Restaurant’s sales have increased? Apply 5% Level of Significance. Solution H0 = There is no any incensement in sales of tea cup H1 = There is an incensement in sales of tea cup Given Population Mean, µ = 500 tea cup per day Number of Sample, n = 12
To be find Sample Mean, ẋ Standard Deviation of Sample , S Solution SN Xi (Xi-ẋ) (Xi-ẋ)2 1 550 2 4 2 570 22 484 3 490 -58 3364 4 615 67 4489 5 505 -43 1849 6 580 32 1024 7 570 22 484 8 460 -88 7744 9 600 52 2704 10 580 32 1024 11 530 -18 324 12 526 -22 484 ƩXi = 6576 Ʃ(Xi-ẋ)2 = 23978 Solution Applying the Formula of t-Test
S= Ʃ(Xi-ẋ)2/(n-1) = 23978/11 t= (548-500)/46.68/ 48/13.49 = 3.558 Solution Applying the Formula of t –Test
By using table of t- distribution
Critical Value = 1.796 Calculated Value = 3.558 1.796 Calculated Value > Critical Value
Conclusion:- 3.558 Null Hypothesis is Rejected, Therefore, There is an increase in sales of tea cup. Independent Two Sample t-Test Formula of Two Mean = X1 sample t- Test Mean = X2 Number of Observation = with Equal Number of Observation = n1 n2 Variance Standard Deviation = S1 Standard Deviation = S2 Question Ajay grows tomatoes in two separate fields. He is curious to know whether the size of tomato plan is differ in both the field or not. He takes random sample from two fields, Here is the summery of Result Mean of sample 1 and Sample 2 is 14 cm and 11 cm respectively. Standard Deviation of sample 1 and Sample 2 is 1.91 cm and 2.45 cm respectively. Number of Plant in Sample 1 and Sample 2 is 7 and 5 Respectively. Solution Hypothesis Formulation H0 = Size of the tomato Plant does not differ in both the Sample H1 = Size of the tomato differs in both the Sample
Given, ● Mean of first Sample, X1 bar = 14 cm ● Mean of Second Sample, X2 bar = 11 cm ● Standard Deviation of First Sample , S1 = 1.91 cm ● Standard Deviation of Second Sample , S2 = 2.45 cm ● Number of Units in First Sample, n1 = 7 ● Number of Units in First Sample, n1 = 5 Calculation of t - Statistics Calculation of t Critical Degree of Freedom = n1 + n2 -2 = 7+5-2 = 10
At 10 Degree of freedom and
2.419 5% Level Significance , t Critical for two tailed Test = -2.228 2.228 +2.228
Conclusion t Calculated > t Critical Null Hypothesis is Rejected, Size of Tomato Plant is differ in both Sample Paired Sample t-Test Formula of Paired Sample t- Test
Where, D bar = Mean of Differences S diff = Standard Deviation of Differences D = Differences (Di= Xi-Yi) n = Number of Observation Question Typing Speed of 9 Students was tested before and after training. State at 5% Level of Significance whether the training was effective from the following Scores.
Student 1 2 3 4 5 6 7 8 9 Before 10 15 9 3 7 12 16 17 4 After 12 17 8 5 6 11 18 20 3 Solution SN X Y D= (X-Y) D2 (D Square) 1 10 12 -2 4 2 15 17 -2 4 3 9 8 1 1 4 3 5 -2 4 5 7 6 1 1 6 12 11 1 1 7 16 18 -2 4 8 17 20 -3 9 9 4 3 1 1 ƩD = -7 ƩD2 = 29 Solution Hypothesis Formulation H0 = There is no any impact of Training on Typing Speed H1 = There is significant impact of Training on Typing Speed
Calculation for Mean of Difference
= ∑D/n = -7/9 = (- 0.778)
Calculation for Degree of Freedom
n-1 = 9-1 = 8 Solution (Cont. …) Now Appling the formula of Standard Deviation of Differences Solution (Cont. …) Calculation of t- Statistics Now from t-Table, we will get Value of t Critical at 5% Significance Level and 8 Degree of Freedom for Two tailed Test
T Critical = 2.306 -2.306 +2.306
-1.362 Chi- Square Test Chi-Square Test The chi-squared test is done in order to check the difference that is present between the observed value and expected value. It was introduced by Karl Pearson in 1900. The Test Statistics do not utilize any estimate like Sample mean, Sample Variance etc. Therefore it is a Non-Parametric Test.
Formula Chi-Square Test =
Question Assembly Election are announced in three state. A Political Party wants to know the proportion of its supporter. Party conducted sample survey of 1000 in each state and found that, there are 300, 350 and 425 Supporter in three states. Using above data, find the proportion of supporter is same or not. Solution ● Given in Question Supporter State 1 State 2 State 3 Yes 300 350 425 No 700 650 575 ● H0 = Supporters across all three states are NOT in different Proportion. ● H1 = Supporters across all three states are in different Proportion. ● Applying Chi-Square Test Observed Frequency Supporter State 1 State 2 State 3 Row Total (RT)
Yes 300 350 425 1075
No 700 650 575 1925
Column Total (CT) 1000 1000 1000 3000
● Expected Frequency = (CT x RT)/ GT CT = RT = GT*** ● CT = Column Total ***This Condition must be ● RT = Row Total fulfill in Chi Square Test ● GT = Grand Total Solution Observed Expected Observed – (Observed – (Observed – Frequency Frequency Expected Expected)2 Expected)2/Expected O E (O-E) (O-E)2 (O-E)2/E 300 358.33 -58.33 3402.38 9.49 350 358.33 -8.33 69.38 0.19 425 358.33 66.67 4444.88 12.40 700 641.67 58.33 3402.38 5.30 650 641.67 8.33 69.38 0.10 575 641.47 -66.47 4418.26 6.88 Solution Ʃ[(O-E)2/E] = 9.49 + 0.19 + 12.40 + 5.30 + 0.10 + 6.88 = 34.36 Calculated Value = 34.36 Now we will find critical Value from Ch-Square Table
Degree of Freedom = (Column -1) x ( Row – 1) = (C-1) x (R-1)
= (3-1) x (2-1) = 2 x 1 = 2 Degree of Freedom = 2
From the Table of Chi- Square, Critical Value at 5% Level of
Significance and 2 Degree of Freedom is 5.991 Critical Value = 5.991
34.36 > 5.991
Calculated Value > Critical Value Therefore, The Null Hypothesis is Rejected Conclusion :- Supporters are Not in same proportion across all three states ANOVA (Analysis of Variance) INTRODUCTION Analysis of variance (ANOVA) is a collection of statistical models and their associated estimation procedures (such as the "variation" among and between groups) used to analyze the differences among group means in a sample.
It is used to compare more than two
population group.
ANOVA was developed by the British
Sir Ronald Aylmer Fisher statistician Ronald Fisher (1890-1962) Question AIIMS, New delhi has conducted a research on effect of Anti Corona drug on three group of people in india belongs to three different states/UT , NCT of Delhi, Kerala and Assam. These states have been chosen due to its geographical location. The drug cure patient in certain hours. The data of patients have been mentioned following.
NCT of Delhi Kerala Assam
2 hr. 4 hr. 2 hr.
4 hr. 3 hr. 4 hr.
3 hr. 7 hr. 4 hr.
7 hr. 3 hr. 10 hr.
4 hr. 3 hr. 5 hr.
Solution = It will solve by One Way ANOVA Given in Question NCT of Kerala Assam Total No of Observation,n = Delhi (X2) (X3) (X1) 15 2 4 2 Formation of Hypothesis 4 3 4 H0 = There is no significant 3 7 4 difference in group Sample 7 3 10
H1 = There is Significant 4 3 5 difference across all groups ∑X1 = 20 ∑X2 = 20 ∑X3 = 25 Solution Step -1: Take the total value of individual item in all sample.
= 62 Solution Step - 5 : Sum of square Total {Xij2 - T 2 /n } = { (4+16+9+49 +16+ 16+ 9+ 49+ 9+ 9+ 4+ 16+ 16+ 100+ 25) - 281.67 } = 347-281.67 = 65.33 Source of Sum of Degree of Mean Square F- Ration Critical Veriation Square Freedom (MS) Value of F SS (d.f)
Between Sample MS Between / 3.88 SS Between/df MS within = From F- 3.3 C-1= 3-1=2 between sample 1.65/5.16 = distributi = 3.3/2 = 1.65 0.319 on Table
Within SS Within/ df N-C=15- Sample 62 3=12 Within sample = 62/12 = 5.16
Total N-1=15- 65.3 1=14 Graphical Representation
Accepted Region Rejection Region
F- Critical F- Calculated 3.88 0.319 Conclusion Since the F value calculated is less then F value critical, therefore Null Hypothesis will be fail to rejected. There is No significant difference in the group of sample.
