 Hello and welcome to this session. In this session we will discuss what leads to misleading confusions from the graphs and we will see how to evaluate published reports and graphs that are based on data. Sometimes, work of a data is misleading and we can draw wrong or invalid confusions. When analyzing graphs or statements of data, we should keep the following things in mind. The sample used for collective data should be selected randomly. It should be unbiased and it should be of appropriate size. Now, the bar graphs and line graphs are misleading if the vertical axis does not start from 0, the scale is not appropriate and if the labeling of axis is not done. For example, see this bar graph. This bar graph shows average house price in the year 1998 and in the year 1999. Now, here you can see average price in the year 1999 is 82,000 pounds and 1998 it is. This in the prices is only that from this bar graph it seems as if the prices tripled in the year 1999 but actually there is a difference of only. You can see that the vertical axis does not start from 0. Now, let us see the improved version of the above graph. Here you can see the vertical axis from 0 and the scale is appropriate. Also, you can see that there is not much difference in the average house prices in two years. Now, show the data in percentages and the total of all the sectors of the pie chart should be equal to 100% and the labeling should be done given the sectors of the pie chart. Now, in published reports and a description of the data we should keep in mind that the analysis of data should not be biased and it should not mess with the people. Now, let us see an illustration which is from Monday to Friday. Which graph could be this leading? Now, here both the graphs represent same data because here the vertical axis is broken difference in the temperature between graph on the left is appropriate display of the data. Now, when evaluating published reports, we should keep in mind the following points regarding the survey and analysis. The type of samples used. Second is the appropriate list of data analysis. The next is the appropriate list of influences and predictions made. Then, next is the validity of conclusions made. Now, let us see an illustration which will explain us how to evaluate a report. Now, a report to published a report based on the survey pie chart and published his report is claimed that 23% lied to Vito's, 16% lie Krispy's. He reported that senior lovers prefer Vito's to both Coral Crunch and Krispy's. Vito's are also higher in vitamins and minerals Now, let us evaluate his report. Now, since a reporter is analyzing failure for sample to be unbiased, the reporter must have taken he has written senior lovers prefer the type of sample is appropriate. Now, let us see appropriateness of data analysis and prediction. Now, the reporter analyzed that the senior lovers prefer Vito's to both Crunchy's and Krispy's. But only 23% like Vito's, 6% that is 22% like equal Crunchy's or Krispy's. So, it is not correct to say that Vito's are mostly prefer by the senior lovers. He has also written that Vito's are minerals and have 14% less sugar. So, it does not seem to have any relevance with the preference of senior. Lastly, we will see the validity of conclusion. Now, the conclusion that senior lovers prefer Vito's is invalid because the data display is misleading. Now, we know that in a pie chart, the sum of sector percentage should be 100% that here the total is 23% plus 16% plus 6% which is equal to 45. It means only 45% like Vito's common branch percent may like some other serial. Thus, the reporter's analysis and his conclusion drawn are absolutely incorrect. Now, the proper display would have been this some other serial which reports and graphs that are based on data and this completes our session. Hope you all have enjoyed the session.