Discuss data presentation in a table andor chart; and explain the components of Discuss the appropriate chart for categorical data based on the variable available in Discuss the presentation of data in a cross tabs frequency distribution table and when S

Including in average are mean, mode, median, strength and weakness of average. While including in dispersion are range, variance, SD; strength and weakness of dispersion. This data can be displayed in graphic such as line chart, hystogram, and box plot. SCENARIO: Case 1. Data Presentation of Categorical Data Look at the skill lab Data appendix Data which have been collected, include several variables: area, age, education level, occupation, exclusive breastfeeding, parity, children age, Hb level , body weight and height. In the previous session, we’ve discuss variables based on their scale of measurement. Understanding the measurement scale of data is important for selecting the way to present the data. . For categorical data, the presentation mainly using three different mode which are table, graphchart and statistics. Table for categorical data are single frequency table, and cross table. Presentation of categorical data in graph can used either bar or pie chart; and statistics that is used are percentage or ratio. Learning Task 1 Discuss the following:

1. Discuss data presentation in a table andor chart; and explain the components of

table and chart.

2. Discuss the appropriate chart for categorical data based on the variable available in

skill lab data 3. Discuss about absolute, relative ad cumulative frequency 4. Below is draft of single frequency distribution table Table-1. The frequency distribution of parity among mothers at Ds. Merdeka, 2011 Parity Frequency Cumulative frequency Relative frequency Cumulative relative frequency 15 1 25 2 30 3 15 4 10 5 or more 5 Total 100 - Please fill the blank cells and Interpret the information on the table above

5. Discuss the presentation of data in a cross tabs frequency distribution table and when

we use this type of table. Give example based on the skill lab data Udayana University Faculty of Medicine, DME 56 6. Below is cross tabulation between area and type of school and nutritional status among teenager in Province Sentosa Table 2. Distribution of teenage nutritional status based on type of school in Province Sentosa in 2012 Type of school Undernourished Normal Obese Total Public 20 a 50 b 15 c 85 Private 10 d 50 e 5 f 65 Total 30 100 20 150 Based on table 2 above, - calculate, the row and column percentages for each cells - Interpret the meaning of row percentage in cell a - Interpret the meaning of column percentage in cell d - Discuss about chart for presenting data in table 2, please provide an illustration.

7. Specific for 2 by 2 cross tabulation four-fold table, we can calculate several numbers

that useful for describing another method for presenting categorical data, which is ratio. The ratio such as: OR Odds Ratio, PR Prevalence Ratio, RR Relative Riskrisk ratio. Discuss and give example based on the skill lab data if possible, or your own example - The Indication of each ratio - How to calculate each ratio - How to interpret the meaning of each ratio 8. There was a cohort study that observed the impact of physical inactivity toward obesity among people age above 40 years old in Province Asri. Underneath is the result after 2 years follow up Table 3. Cross table between physical activity level and the obesity among people age above 40 years old Physical activity Obese Normal Total Less active 30 25 55 Active 10 35 45 Total 40 60 100 From table 3, - Determine the type of ratio that appropriate for comparing the risk of obesity based on degree of physical activity? - Calculate and interpret the meaning of the ratio Udayana University Faculty of Medicine, DME 57 Case 2. Data DescriptionInterpretation of Continuous data Re-open the Skill lab data appendix. Previously, you have discussed about data presentation of categorical data. Similar to categorical data, the mode of presentation for continuous data divides into three ways namely; 1 Table, 2 Graphs and 3 Statistics. The tables that can be used are single frequency tables when the range of the data is narrow; on the other hand, using grouped frequency table when the range of data is wide. Several types of graph can be used for presenting continuous data include histogram, box plot, steam and Leaf plot, polygon and scatter plot. Meanwhile, statistics for continuous data comprise measure of central tendency and Measure of dispersion. The measure of central tendency is the value where the data are concentrated. It includes mean, median and modus. Meanwhile, measure of dispersion show the spread of the data includes range, variance, standard deviation, and inter quartile range. It is also important to understand the distribution of data which can be seen by plotting them on a histogram. The distribution will guide us to select the appropriate statistics to represent the data. Learning Task 2 Discuss the following 1. Discuss and explain the definition, of central tendency mean, modus, median. 2. Discuss the distribution of data which can be determine by the value of mean, median and modus of the data 3. Case 1; If from 100 samples, we found mean Hb level is 12.8 mgdl, the median is 11.0 mgdl and the mode is 11.0 mgdl. - Is the data tend to distribute normally or skewed? - Which measure of central tendency is appropriate for describing the data? 4. Discuss and explain the definition of measure of dispersionspread range, variance, SD, percentile, quartile and inter quartile range - Based on case 1 above, what measure of spread is appropriate for describe the data? 5. Based on continuous variables on skill lab data, discuss the appropriate chart that can be used for data presentation and provide illustration 6. In the skill lab data, we have data on children’s age and body weight. Suppose we wish to show both data in a graphto show the relationship between both variables, what type of graph can be used and make an illustration Self Assessments: Udayana University Faculty of Medicine, DME 58 1. Explain the components of a one-way frequency distribution table 2. How can you conclude the content of a frequency distribution table? 3. Explain the application of cumulative percentage 4. Cumulative percentage is applied to which kind of data? What is it for? 5. Explain how to conclude the content of a cross tabulation table 6. Explain briefly the definition of: column, row, and total percentages 7. Explain some indicators to show the relationship regarding to various categories of variables 8. Explain the understanding and application of central tendency average: mean, median, and mode. 9. Explain the understanding and application of dispersion spread: range, variance, and SD. 10. Explain the advantages and disadvantages of central tendency average: mean, median, and mode. 11. Explain the advantages and disadvantages of dispersion: range, variance, SD. 12. Explain the differences of data presentation and data description of age and formal education, before and after the categorization. 13. Explain the differences of data presentation and data description of body height, body weight, and Body Mass Index BMI. Udayana University Faculty of Medicine, DME 59 M O D U L E ~ 8 SKILL LAB II Skill Lab Manual, Kirkwood Sterne Data Presentation and Data Interpretation dr. Putu Ayu Swandewi, MPH dr. Gede Artawan Eka Putra, M.Epid AIMS: To demonstrate ability to search, organize and interpret informationdata from different sources in order to assist in diagnostic, therapeutic and health. LEARNING OUTCOMES: 1. To present data into some forms for categorical and continues data 2. To interpret data accurately CURRICULUM CONTENTS: 1. Data presentation for categorical data 2. Data presentation for continues data SCENARIO LEARNING TASKS Refer to Cases of Data presentation and Data Description DAY 9. Udayana University Faculty of Medicine, DME 60 M O D U L E ~ 9 Reference Kirkwood Sterne, Chap. 5-8, 10, 14, 15 Inferential Analysis and Interpretation of Analysis Results Hypothesis Test Putu Ayu Swandewi, MD, MPH Gede Artawan Eka Putra, MD, M.Epid AIM: Demonstrate ability to search, collect, organize and interpret informationdata from different sources in order to assist in diagnostic, therapeutic and health. LEARNING OUTCOME: Be able to interpret inferential and interpretation of result analysis CURRICULUM CONTENTS: Inferential Analysis and Interpretation of Analysis Results Hypothesis Test ABSTRACT THEORETICAL DISTRIBUTIONS, CONFIDENCE INTERVAL, AREA UNDER NORMAL DISTRIBUTION, AND HYPOTESIS TEST For statistical test purpose, we must have basic understanding on theoritical distribution, confidence interval CI, one-tail and two-tail test. In a theoritical distribution there are several topics need to be understood, are the probability distribution, characteristics of normal distribution, area under normal distribution curve, definition of Standard Normal Deviate, and also definition and the application of CI. The important point is the understanding on whether a certain data is normally distributed or not. Parametric test require normal data distribution . Related with normality of data, extreme value outliers may influence data normality. By deleting those values if the sample size is sufficient, we may produce data with normal distribution. We also can replace those values with the median to gain normal distribution data. If those efforts are not succeed, we may use transformation Udayana University Faculty of Medicine, DME 61 methods based on power of transformation in the Explore. If the data still not normally distributed, we may choose non parametric test. In a hypothetic test, we must know the asumptions used, definition of nul and alternative hyphotesis, relation between hyphotesis test with probability p, confidence interval CI, the errors in hyphotetis test, and the application of one and two tail tests. SCENARIO: Case 1. Theoretical Distributions, Confidence Interval, Area under Normal Distribution For statistical tests, it is important to understand probability theory, theoretical distributions, confidence intervals CI, one-tail and two-tail statistical tests, and error in interpreting hypothesis tests. Learning Task 1 Discuss the following 1. Explain the definition of probability, marginal probability, conditional probability and joint probability 2. Explain the definition of “the addition rule and multiplication rule” 3. Table 1. Cross tabulation between economic status and nutritional status of children under 5 years in Desa Amerta 2012 Economic status Under nourish Normal Obese Total poor 30 35 5 70 affluent 10 55 15 80 Total 30 80 20 150 Based on table above, calculate the probability: a. What is the probability if we selected one child randomly from Desa Amerta, will be an under nourish kids in? What type of probability it is? b. What is the probability of under-nourish, if the kids from poor family? What type of probability it is? c. Probability of a child who is randomly selected from this group will be from poor family or an obese kid? d. Probability of a child who is randomly selected from this group will be from affluent family and is obese 4. Describe the type of probability distribution of a random variable. Give examples 5. Describe the characteristic of normal distribution and the use of normal distribution? Udayana University Faculty of Medicine, DME 62 6. Distinguish the difference between “normal distributionGaussian distribution” and “standard normal distribution” 7. Describe about z score and its use? 8. If from a study we found the mean of Hb level from 200 samples of children is 11.8 mgdl, and the standard deviation is 0.5mgdl. It is assumed that data was normally distributed. Based on the used of normal distribution and the area under normal curve z score a. What is the probability of children in the population has Hb level 12mgdl b. What is the probability of children in the population has Hb level 11 mgdl 9. Describe about sampling distribution, standar error of the mean and confidence interval CI of the mean 10. Based on case 2 above: - Calculate the standard error of the mean - Calculate the 95 CI of the mean - Interpret the 95 CI of the mean Case 2 Hypothesis Testing and Linear Correlation Re-open skill lab data appendix. Recall that the type of data presentation is based on scale measurement and also the type of hypothesis testing will be based on the scale of measurement and purpose of the analysis. Basically, there are two types of hypothesis testing; namely: testing the difference and testing the association. The basic concept of hypothesis testing should be understood before performing the test. In hypothesis testing we are testing the possibility of the difference or the association is likely to be true in the population. There two type of hypothesis, null hypothesis Ho and alternative hypothesis Ha For testing the association between two continuous variables we can use correlation test with correlation coefficient from Pearson r. In the correlation test, we can determine the strength and direction of the association. Correlation may be a linear and non-linear, symmetric and non-symmetric, straight and reverse correlation, and relational and causal. In this case, we will only discuss linear correlation. Learning Task 2 1. Discuss the indication of linear correlation analysis. 2. From the variables in skill lab data, discuss example of possible correlation analysis 3. Discuss the two ways that we can use for determining correlation between two variables? 4. Discuss the use of coefficient correlation? 5. Discuss the meaning of negative and positive correlation, provide example Udayana University Faculty of Medicine, DME 63 6. Discuss the strength of the association, when it is considered low, moderate, high and perfect, describe with example Self Assessments: 1. Draw figure of t-distribution, chi-square  2 distribution, F-distribution, Poisson distribution, Log normal distribution, Binomial distribution. What are the differences of those six distributions? 2. Explain the probability distribution of a random variable 3. Explain the definition of Normal distribution Gaussian distribution. 4. Explain the characteristics of Normal distribution. 5. What is the total area under the normal curve between: - and +, -1.96 and +1.96, -2.58 and +2.58. 6. Describe the definition of SND Standard Normal Deviate. 7. Explain one-tail and two-tail statistical tests and when they can be applied. 8. Describe the strength and direction of quantitative variables correlation. 9. What does it mean if two quantitative variables have correlation coefficients r which are negative or positive? 10. What does it mean if two quantitative variables have correlation coefficients r of 0 or 1? Udayana University Faculty of Medicine, DME 64 M O D U L E ~ 10 Reference Kirkwood Sterne, Chap. 7, 9, 11, 16, 18, 19 Significance Test for Categorical and Interval Data dr. Putu Ayu Swandewi, MPH dr. Gede Artawan Eka Putra, M.Epid AIM: Demonstrate ability to search, collect, organize and interpret informationdata from different sources in order to assist in diagnostic, therapeutic and health. LEARNING OUTCOME: Understand the significance test for categorical and interval data CURRICULUM CONTENTS: Significance Test for Categorical and Interval Data ABSTRACT

1. CATEGORICAL DATA