• Identify data source. You need to have at least 100 data points. If you are going to use other people’s data, make sure the source of data is clearly identified. • You may already have data you can use in the form of historical data such as utility bills, phone bills, bank statements, etc. • If you have a job, you may be able to use data from work such as daily scrap figures, product sales, product returns, customer complaints, etc. You may be able to observe something at work to collect the data such as the number and types of pizzas ordered during peak hours or the type of merchandise returned by customers for credit. • You may already have the data available if you can get access to it. Don’t forget to collect relevant other information with your samples that may become an important part of the analysis (e.g., time of day, person involved, machine used, shift, supplier, etc.). • Your data should fit into an Excel spreadsheet with columns representing each variable or time of information in the sample (e.g., date, time, number, cost, etc.) and each row of data representing one observation or sample. So, if you take 100 samples you would have 100 rows of data. Statistical software likes this format. Avoid using a “grid” with two axes of data–such as (breakfast, lunch, dinner on the y-axis and date on the x-axis.