Welcome to lecture 25 about the data collection you can see the corresponding B okay reference on the screen. The purpose of this lecture is to provide you with concepts of continuous and discrete data as well as provide understanding about various measuring scales such as nominal ordinal, interval and ratio scales. At the end, we will discuss about various data collection methods. It's accuracy and integrity. We had briefly discussed about data in the statistics lecture. As we are aware data is required of actual observations.

For example, for a cricketer it means runs, batting, average wickets etc. For a quality professional data means the rejection rate reworks output specifications etc. For a manufacturer data means sales figure, productivity, inventory level, etc data and facts are the backbone of Every six sigma projects without facts, the problem solving efforts are reduced to just a guessing game. The chances of success will be low without an analysis of data. Basically, data is classified into two types variable our continuous data attribute or discrete data. Variable Data are those that can take fractional values such as length, breadth, diameter, time, etc.

The attribute data The outcome of sum count for example, number of customers per day number of telephone calls received in a week, rejections per hour absentees or a month etc all these data cannot be in fractions it will be n counts that is learn about the measurement scales. Now, for this purpose let us visualize different situation such as a gear shaft producer likes to check his production quality the measurement out We'll be something similar to diameter and mmm 19.1 18.5 and so on. A manager at three star hotel will be interested in number of guests checked in per day such as Android 110 etc etc. A customer relations manager at a telecom company analyzing the customer feedback categories, such as poor, average, good, excellent, etc. A government official assigned to conduct survey on marital status of population single married, divorced, widower, different situations of measurement, isn't it?

Naturally, there should be a classification of measurement for the above set of data. All measurements in science could only be conducted using four different types of scales namely ratio, interval, nominal and ordinal. In ratio scales, there will be a meaningful difference between the measurements. Also, there exist an absolute zero Examples are length, age, weight, etc. In interval scales also there will be a meaningful difference between the measurement scales. However, there will not be an absolute zero.

The examples are date, temperature, etc. in nominal scale it's mainly for categorical data. For example, person names, designation status, different machines, etc. Ordinal Scale is similar to nominal scale, but there would be a meaningful order Between the data sets for example, different grades, satisfaction score, such as poor, good, average, excellent, etc. What are the sources of data data could be obtained by observations, experiments survey or the historical data stored in databases. However, during the data collection process, it is essential to ensure that the data collected is not time or person dependent as well as ensure data integrity, data precision and accuracy.

That means using the right operational definition and appropriate gauges for collecting data, data consistency by ensuring there are no changes in measurement system, operational definitions, metrics etc. During the course of project time traceability data must be traceable to the time it was collected. data can be collected in numerous ways using check sheets. Universal data sheets, coded data sheets etc. However, a good data collection plan is essential to bring down meaningful data for the project. A sample data collection plan could be as shown on the screen.

Could you see the operational definition type of data sizes of data, how to be collected and by whom it will be collected etc are well explained in this plan. It is also important to assure the accuracy of data. Accuracy could be assured by sampling method sampling method is a process of collecting a portion of data for study from population. different strategies of sampling are random sampling, simple random sampling, where a population is homogeneous, each item for sample have equal chances of being selected. can use random table, ball of numbered chit etc. for selection of item stratified sampling when population is known to be from different sources, machines Production shifts, operators etc items are selected from each group judgmental sampling we are sampling is done based on once experience and opinion that's all for this lecture we have covered all the points as per the B Okay.

Now, let us proceed to the next lecture on descriptive statistics. Thank you