Clarify the sampling objects and population

Samples must be drawn independently and randomly

The sample drawn must be representative

The sample should be appropriately balanced

After the sample extraction plan is determined, the sampling cannot be expanded at will.

Simple random sampling: refers to selecting several sampling units from the population to form a sample, and ensuring that all possible combinations of n sampling units have an equal probability of being selected.

Systematic random sampling (mechanical sampling, equidistant sampling): a sampling method that arranges the products to be sampled according to a certain symbol and randomly selects one or a group of products according to a set of rules.

Stratified random sampling: Divide the overall unit into non-overlapping layers according to certain important signs, use simple random sampling or other sampling methods to select several sample individuals in each layer, and form a sample from the sample individuals in each layer

Cluster random sampling: Divide the population into several non-overlapping groups, each group consists of several individuals, then select several groups from the population, and all the individuals in the extracted groups form a sample

Determine the specific purpose for which the data is collected

Determine what needs to be recorded on the form

Prepare forms for recording information

Try out the form by collecting and recording some information

Review and revise forms as necessary

Use brainstorming to guide group members to think creatively and generate or clarify a large number of ideas, questions or issues.

Brainstorming can be used to identify quality problems existing in the process and seek solutions, and can also be used to identify potential quality improvement opportunities.

This method is commonly used when drawing cause and effect diagrams, tree diagrams, and affinity diagrams.

Preparation Phase

creative thinking stage

finishing stage

When layering, the choice of layering logo is very important

According to the purpose of stratification, it should be distinguished according to certain signs, so that the fluctuation range of data within the same level is as small as possible, and the difference between layers is as large as possible, otherwise the role of classification and summary will not be achieved.

Usually people, machines, materials, methods, environment, time, etc. are used as symbols of layering.

clear object

Data collection

layered logo

Organize data

Classified by layer

Draw a hierarchical classification chart

Comparative Analysis and Final Inferences