deviation Inferential analytical statistics methods enabling a conclusion t be

# Deviation inferential analytical statistics methods

This preview shows page 5 - 7 out of 25 pages.

deviation) Inferential (analytical) statistics: methods enabling a conclusion t be drawn from data (correlation and regression). WHY SAMPLES? Populations are generally too big and/or individual items too inaccessible to be examined entirely. Examining whole populations can be too time consuming and costly. The full extent of a population may not be known (people with unsuspected diabetes, victims of crime). Examination may result in the destruction of items (testing flammability). Stages in sample data collection: define problem (define population), design sample (sample size and method) Draft questionnaire (conduct pilot survey), collection and check data (code responses for tabulation) organize, analyze and interpret data (tabulate, graph), report findings (recommend a course of action). SELECTION METHODS : (sampling is an entire process which starts with defining a population and ends with drawing a conclusion about the population. The ways in which a sample may be drawn are just on stage in the process-the selection method). RANDOM SELECTION : (if conclusions relating to the whole population are to be drawn, samples must be free of bias). A random sample is selected in such a way that all items have an equal chance of being included in the sample (using random number tables or random number generator). To use this method: the number of items in the population must be known, it must be possible to match each item in the population uniquely against each random number generated, ranom numbers should not be duplicated. Simple random sampling is most appropriate when the entire population from which the sample is taken is homogeneous. Because of the time it takes to sort the random numbers and match each to an item quasi-random methods are often used in practice. Systematic Selection : This quasi-random method uses a constant interval between items selected from a random start. It is also called interval sampling. The interval may be: a number of items or a monetary value.
6 Illu.2. Systematic selection- a number of items: A sample of 100 invoices is to be selected from the 5000 invoices , them sample size would be 5000/100=50. Start at a random point between 1-50, and go on adding 50 to the number selected. 3. Systematic selection-a monetary value (based on monetary value is also called cumulative monetary amount or CMA selection. It is widely used in audit sampling). A sample of 100 is to be selected from the 5000, range in value from 10- 1200. A higher proportion of the monetary value of the bills will be examined if a value-weighted selection is made from the monetary value of the population which is 2150000, Interval=2150000/100=21500. Start any random point between 1-21500. Sum up in order. The first bill selected will be the one which takes the cumulative amount to the next one selected.

#### You've reached the end of your free preview.

Want to read all 25 pages?

• Spring '16
• amarnath
• management accountant