ICGB Sample Questions & Answers

R-squared, also known as the coefficient of determination, measures the proportion of the variance in the dependent variable (Y) that is predictable from the independent variable(s) (X). An R-sq of 85% means that 85% of the observed variation in the number of defects is explained by the linear relationship with the number of lines of code. It does not imply causation.

The principle of 'garbage in, garbage out' is critical in Six Sigma. Before any process analysis can be trusted, the measurement system itself must be validated. The lack of synchronized systems and manual entry points to significant potential measurement error. The most critical step is to establish clear operational definitions for exactly what 'Order-to-Delivery Time' and its components mean, and then perform an MSA (like a Gage R&R study, modified for a transactional process) to understand and quantify the error in the data. Without this, any conclusions drawn in the Analyze phase would be invalid.

A Pareto Chart is a bar chart where the bars are ordered from highest frequency to lowest, illustrating the 80/20 rule (Pareto Principle). This principle states that for many events, roughly 80% of the effects come from 20% of the causes. The chart visually separates the 'vital few' causes that contribute most to the problem from the 'trivial many', allowing teams to focus their efforts effectively.

Internal Failure Costs are incurred when defects are discovered before the product or service is delivered to the customer. Examples include the cost of rework, scrapping defective products, downgrading products, and machine downtime caused by defects. These are costs of nonconformance found within the organization.

This is a Control Method, specifically a prevention-type Poka-Yoke. It physically prevents the error from occurring. Warning methods would only signal that an error has been made (e.g., a light or buzzer), but would not stop it from happening. The asymmetric pin design makes it impossible to assemble the part incorrectly, thus preventing the defect at the source.

Ordinal data, like a 1-5 satisfaction scale, is non-parametric. A 2-sample t-test and ANOVA are parametric tests that assume continuous data and normality. The Mann-Whitney test is the appropriate non-parametric alternative to the 2-sample t-test for comparing two independent groups when the data is ordinal or not normally distributed. It compares the medians of the two groups.

A strong problem statement must be measurable. 'Too low' is subjective and does not quantify the current level of performance or the gap between the current state and the desired state. It should include data, for example: 'The average customer satisfaction score has dropped from 4.5 to 3.2 over the last six months.'

'Customer satisfaction' is a very broad concept. A good problem statement should be specific. Is the dissatisfaction related to product quality, delivery time, customer service, or something else? A focused problem statement like 'Complaints regarding late deliveries have increased by 50% in Q3' is much more effective.

General guidelines for Gage R&R % Study Variation are:

• Under 10%: Acceptable.
• 10% to 30%: Marginally acceptable or conditionally acceptable. The decision depends on factors like the application's importance, cost of the measurement device, and cost of repairs.
• Over 30%: Unacceptable; the measurement system needs improvement.
  A value of 25% falls into the 'marginally acceptable' category.

In the Y = f(x) relationship, Y is the output or effect (the problem being measured), and the 'x's are the inputs or causes that influence the output. The number of invoice errors (Y) is a function of various inputs (x's). The level of training is a potential input or cause (an 'x') that could directly impact the error rate. The other options are all different ways of measuring the output (Y) or consequences of the output.

This statement is true. Cp measures the potential capability of a process, assuming it is perfectly centered. It compares the process spread to the specification spread. Cpk measures the actual capability, taking the process centering into account. A large gap between a high Cp and a low Cpk means the process has low variation (good potential) but is not centered correctly within the specification limits, causing defects on one side.