Data Presentation: Tables & Graphs - PDF

Data presentation: tables and graphs Dr Asmaa Fawzy Lecturer of Public Health, Faculty of Medicine Cairo University The aim of presentation is to find-out the common finding, The extent in which studied group differs in values (group variation), To identify the strange (odd)values...

Data presentation: tables and graphs Dr Asmaa Fawzy Lecturer of Public Health, Faculty of Medicine Cairo University The aim of presentation is to find-out the common finding, The extent in which studied group differs in values (group variation), To identify the strange (odd)values in the group. Methods for presenting data ❖Tables: Show data details in an arranged and grouped way. ❖ Graphs: Show simple impression of the data. A good table has: Title explaining its contents Meaningful percent from row or column Suitable number of rows (4-12) avoid rows with zero frequency Clear and self explained without reference to the text Headings of different columns should be clearly defined with the units of measurement. Meaningful total to saves the reader the trouble of summation Not everything displayed in the table needs to be mentioned in the text Round numbers to one or two decimal places.  A good table Sex distribution of medical statistics course students (Kasr El-Aini 1998) Title &Abbreviation Number Percent Sex Heading of column& Measurement unit Males 11 61.1 Females 7 38.9 Total 18 100.0 Meaningful% from Raw or column C&R arrangement Clear self explained 1. Spread sheet ⚫When you finish collecting data, you need to know what it says. ⚫To be able to do that we have to see all the data together ⚫The first step is to spread data items for each individual in an arranged in the same way, this is known as the spread sheet or the raw data table. ⚫It shows all the data we need but it is difficult to make outconclusion at a glance. a spreadsheet of 18 of medical statistics course students (Kasr El-Aini 1998).  No Sex Age Result Degree 1 Male 35 Fail 50 2 Female 30 Pass 70 3 Male 40 Pass 90 4 Male 45 Pass 80 5 Female 33 Fail 40 6 Female 34 Fail 55 7 Female 32 Pass 60 8 female 28 Fail 50 9 male 25 Fail 45 10 male 30 Pass 70 11 male 40 Pass 72 12 male 42 Pass 75 13 male 45 Pass 80 14 male 41 Fail 55 15 female 55 Pass 66 16 female 44 Pass 67 17 male 36 Pass 90 2.Simple table showing single variable: Ǫualitative variable Ǫuantitative variable Nominal or ordinal Discrete or continuous Limited number of categories Many possible values so we need to combine categories  Sex distribution of medical statistics course students (Kasr El-Aini 1998) Number (frequency) Percent Sex (relative frequency) Males 11 61.1 Females 7 38.9 Total 18 100.0 Age distribution of medical statistics course students (Kasr El-Eini 1998). Frequency Age group 25- 2 30- 5 35 - 3 40 - 5 45 - 2 50- 0 55 1 Total 18 3.Contingency tables or cross tabulation of two variables: Used to analyze the relationship between 2 or more variables. Most usually categorical  Frequency distribution of exam results according to students' sex Fail Pass Total Result SEX Sex Male 4 7 11 Sex Female 3 4 7 Total Exam 7 11 18 result Results You can comment on: 1-Each variable separately by looking at the totals 2-Examination results for males only and for females. Only.  Percent distribution of patients with IHD according to sex IHD Not IHD Total N % N % N % Male 20 57 15 43 35 100 Female 10 25 30 75 40 100 Total 30 40 45 60 75 100 You can comment on: 1-Each variable separately by looking at the totals 2-Examination results for males only and for females. Only. One picture is more valuable than a thousand words! ❑Graphs are more capable of gaining attention, stressing a certain phenomenon and giving a quick idea about the general situation. ❑Graphs are not substitutes for tables as they use approximate values. ❑graphs should be self explained, so that the contents should be understood without reference to the text.  Any graph should have: 1. Title : preferred to be below the graph 2. Legend: a key for the pattern and the level it express should be shown at the side of the graph. Graphic presentation of data Ǫualitative data Ǫuantitative data Graphs for qualitative data 1. Pie chart (circular chart) 2. Bar chart 1.pie chart(circular chart) It consist of a circle whose area represents the total frequency, which is divided into segments which represent the proportional composition of the total frequency. 36.10% yes no 63.90% Hypertension prevalence among studied population 2- Bar chart: Categories are listed horizontally in systematic order It is a graphical presentation of frequencies or relative frequencies (percentages) by rectangles of constant width drawn with lengths proportional to the frequency or relative frequency. A space separate each bar to emphasize the nominal or ordinal nature of variable.  100.00% 12 10 80.00% 8 60.00% third Number 6 4 40.00% fourth 2 0 20.00% Male Female Sex 0.00% yes no 120 100 80 60 40 no yes 20 0 Experienc of students with teaching methods  Bar charts could make it easier to compare uni- variate distribution for two or more groups. 100% 80% 60% severe 40% moderate mild 20% 0% male female Disease severity among males and females  Bare charts are suitable to present data when the total is more than 100% as in case of having more than one response from interviewed respondents. 50 45 40 30 method of ad 30 20 daily work 20 15 one tablet 10 no felt side effects 0 views Percent of patients according to their view to tested drug Graphs for quantitative data 1. Histogram 2. Frequency polygon 3. Frequency curve 4. Line graph 5. Scatter dot diagram  1- Histogram : It is graphical presentation of a frequency distribution in which rectangles proportional in area to the class frequency are erected on the horizontal axis Each histogram has a total area of 100% 1- Histogram(cont.): Histogram is similar to bar chart of qualitative data but the difference is that the columns in histogram are adjacent to each other while bar chart is separate. As in bar chart each column show different group, while histogram shows a continuous variable with the beginning of each group immediate to the end of the preceding one.  12 10 8 Number 6 4 2 0 Male Female Sex Quantitative data is continuous, Qualitative data expresses so, we can link it to each separate groups. other by connected bars. 1- Histogram(cont.): For discrete variables (e.g. number of children per family), the number representing the value should be centered below each bar to emphasize the discrete nature of the variable or we can use bar chart. 2- Frequency polygon: Derived from the histogram by connecting the midpoints of the tops of the rectangles in the histogram. 3- frequency curve: The smooth line that passes between the midpoints of the tops of the rectangles in the histogram. 4.Line graph Useful for numeric data if you want to show trend over time. 4- Scatter dot diagram (dot graph) It is a diagrammatic presentation of the relation between two quantitative variables. Each observation is represented by a point corresponding to its value on each axis (two variables for one individual). Thank You 34