The following are posts from forum conversations, please reply to all three individually1-What is sampling and why do we need sampling?Sampling is a way summarize population information or characteristics

Subject:AccountingPrice: Bought3

Share With

The following are posts from forum conversations, please reply to all three individually1-What is sampling and why do we need sampling?Sampling is a way summarize population information or characteristics. Often, the population is spread out, sometimes geographically, and can be difficult to gather all information needed for analysis. Sampling is a way to take a portion of the population, or data set, and infer representative characteristics of the entire population. (Albright & Winston, 2015)Sampling offers many benefits for discovering characteristics of a population set, these include reduced cost, quicker and more efficient, greater scope and flexibility and greater accuracy (Chambers & Clark, 2012).What are the common sampling methods?There are many sampling methods. The most common of these (and used for throughout our text) is simple random sampling. Simple random sampling is the method of choosing a sample size n to represent a population N; the sample size n has the same probability of every other sample size n of being chosen. (Albright & Winston, 2015)Another method for choosing a sample is systematic sampling. Systematic sampling is a way of choosing a random sample through a set process. An example of this type of sampling would be to choose demographics from bank accounts from a population listed by account number. As long as the account numbers are generated randomly, then choosing every 5th (or nth) account would give you a systematic sample. The issue with this method is that the data needs to be arranged in no particular order. Meaning bank accounts cannot be issued based on income or any other specific characteristic. If bank accounts are issued based on income levels, for example all customers with an income over $100,000 receives an account beginning with a 5, then the systematic sampling would be biased and not a good method for true random sampling. (Albright & Winston, 2015)Stratified sampling is a method that allows the analyst to choose population sets according to a particular strata. The population is divided up into subsets called strata and each strata is divided based on similar characteristics. Then, random samples are taken of each strata proportionately and combined for the entire same size, n. (Albright & Winston, 2015)Cluster sampling is a way of sampling populations based on concentrated areas of the population. This is easiest to see in cities or towns where there are literally population clusters in which to choose the sample. The largest drawback to this method is that the clusters may not be a true representation of the entire population, but the convenience of this method outweighs that statistical detail. (Albright & Winston, 2015)Multistage sampling is a more realistic sampling method that is used in more complex, real-world applications. In this method, there is one method used in the first stage of the sampling, followed up by a different method for a subsequent stage, and so on until the sample size is attained. The textbook example of this is polling of people where cluster sampling is done in the first stage, then systematic sampling within the cluster. (Albright & Winston, 2015)Could you provide a real world example to illustrate which method is more appropriate to choose in reality?There are many examples of random sampling that most of us have encountered. These include mall polls, restaurant polls, online polls and political polls. These are generally common is areas with larger populations and can be seemingly random. (Random Sampling, 2016)What are your criteria?In these sampling sets, the criteria really can be defined as clusters of people and those that are willing to take the time for the survey. Criteria are still defined according to the defined problem, population, and overall goals of the survey. For example, a survey could be taken at a local to determine the average income of the mall’s customers. In this respect, the survey would be quite relevant and if people are willing to check a box next to their income level, then the mall management can better understand the income levels of their customers. (Random Sampling, 2016)2- Can you give me a working example of sampling? Maybe a way you would use sampling in your job? What are the benefits to using sampling? Any drawbacks?Share your thoughts.3- Hypothesis testing is a way to determine which of two competing hypothesis is the best decision. In hypothesis testing, there are two hypotheses, the alternative and the null. The null hypothesis is simply the status quo. One way to remember this is that nothing changes, so a “null” change. The alternative is the hypothesis that the one that an analyst would try to prove. Something that would break from the “norm.” This is also called the research hypothesis. (Albright & Winston, 2015)When attempting to prove an alternative hypothesis, there are two types of errors that an analyst can make: Type I and Type II. The type I error occurs when a null hypothesis is rejected, but it is actually true. The type II error occurs when a null hypothesis is accepted, but it is actually false. (Albright & Winston, 2015)Type I errors are generally more costly. This makes complete sense. If you sink a significant amount of money into a new product and that product has no demand, then you are back to the status quo. The money that was spent on the new product then is essentially lost. However, if you stick with the status quo and never create any new products, then you could be losing money by expecting your demand to stay high. This would be a type II error. Because type I errors are costly, they also lead to conservative decision making. (Albright & Winston, 2015)Hypothesis testing can be done in an unlimited number of scenarios. Philip Sedgwick discusses the pitfalls of hypothesis testing when it is done multiple times over in a study (2014). The study set out to determine if home based intervention for parents with infants (children under 2) would reduce the children’s BMI. The null hypothesis then would be that intervention will not decrease infants’ BMI and the alternative hypothesis was that intervention will decrease the infants’ BMI. The study actually set out to determine 24 different hypothesis, but the aforementioned hypothesis was the overall study purpose. (Sedgwick, 2014)