Business Project

The population of Alaska is 736,732 (United States Census Bureau, 2016). This gives Company A a tremendous sample to see whether outsourcing or insourcing is better for their business, which is what the company would like to determine.

Target population and brief reasoning

The target population is other farmers in the state of Alaska simply because they are the target customers.

Data Mining

The samples provide data from which the analysis is determined. Samples conducted through data mining involve the costs incurred by other companies to either produce the tomato seeds or purchase them. Most farmers are getting seeds from outsourced locations, so this will provide much of the data. Data mining tools such as the application of artificial intelligence will be applied (Arazi, et.al, 2012), but gathering data could potentially prove to be labor intensive. Such methods will aid in collecting data concerning the costs of production and compare them to the price that other farms are paying for outsourcing the same kind of seeds.

Sample size

For the purpose of this company the sample size is 385, with the population proportion unknown (see Appendix A).

Method of random sampling

Stratified random sampling serves as the most reliable method of sampling to represent each group of a sample. According to Lind, Marchal, and Wathen, a stratified random sample occurs when, “A population is divided into subgroups, called strata, and a sample is randomly selected from each stratum” (2015, p. 252). So when comparing expenditures and profits experienced by other farms and seed manufacturers in the industry against those of Company A, stratified sampling provides the soundest method.

Validity and Reliability

The company intends to demonstrate the validity and reliability of findings through testing of the hypothesis statement: Despite increased overhead and expenses (such as fuel costs per gallon), it is exponentially more beneficial to grow seeds for use in Alaskan climate than to purchase them from current sources. Successful execution of hypothesis statement testing serves to prove the validity and reliability of the hypothesis statement. The company will execute the six steps of hypothesis testing to prove this point.

Part 2: Data collection plan and conclusion

The data gathered will be accumulated through observation and documentation. Observation is a way of collecting data by watching specific behaviors, events, or noting characteristics in a natural environment. Two rounds of observations will be used. Using methods like direct and indirect observations will help clean the data consumed.

Direct observation is real-time observation. For example, the growth of Alaskan tomato crops can be observed and documented exactly as observed. Indirect observation is observing and documenting the results of a process that has occurred. For example, one can measure the amount of tomato seeds harvested and sold for profit at the end of a cycle. This will be the second round of observations and can provide another filter to analyze our data collected.

Data will be stored in various locations utilizing multiple platforms. Data will be stored on computers and backed up on our company’s database. The database will be secured on our premises, encrypted, and have limited access. In addition, information will be backed up through cloud storage, which could be beneficial for disaster or corruption security.

Levine, P. A., & Limberg, A. L. (2014). U.S. Patent No. 4,437,764. Washington, DC: U.S. Patent and Trademark Office.

Appendix A: Calculation for sample size

where:

n = size of sample

z = desired level of confidence standard normal value

= population proportion – since the number is unknown, .5 is used