Christina Walters, vice president of finance for Underwood Company, attended a seminar three years ago on activity-based costing. She immediately began a project to gather data on activities for three plants that produced the same products and were similar in size and organization. After the first three months, she had sufficient data to convince herself that the consumption ratios of projects differed significantly. Furthermore, the nonunit-based overhead costs represented about 47% of the total overhead costs. Based on this input, she decided to implement ABC in the three plants on a pilot basis before expanding the system to the company's remaining 47 plants. She wanted at least two years' experience with the system before committing the company's resources to a major new information system. To prepare her presentation for the rest of management, Christina decided to calculate two overhead cost formulas for the plant: one using direct labor hours as the only driver and a second using direct labor hours and other drivers for which information had been gathered over the past 18 months (in three plants). Pooling monthly data for the three plants pro¬

duced 108 data points. Computer printouts for the two equations are presented below:

Required:
1. Explain how regression analysis can help Christina decide whether to implement ABC for the remaining plants.
2. Based on the two regression equations, evaluate the decision to use ABC for the three plants. Do the results suggest that any further actions are needed?
3. After evaluating the results, Christina decided to run regressions for individual activi¬
ties, including some with activity drivers that had not yet been used. Why would it be useful to know cost formulas for individual activities?

Transcribed Image Text:

Parameter Estimate t for Ho Parameter = 0 Prt Standard Error of Parameter Intercept 80,000 2.326 0.045 Direct labor hours 1.96 9.956 0.000 34,394 0.197 R Square (R2) 0.46 Standard Error (S.) 3,000 Observations 108 t for Ho Parameter Estimate Parameter = 0 Prt Standard Error of Parameter Intercept 22,000 1.96 0.0250 11,224.490 Direct labor hours 1.90 8.67 0.0001 0.219 Number of moves 100 12.50 0.0001 8.000 Setup hours 50 81.96 0.0001 0.610 Purchase orders 200 0.85 0.2000 235.294 Multiple R Square (R) 0.76 Standard Error (S,) 2,500 Observations 108