| Model Coefficients | ||||
| term | estimate | std.error | statistic | p.value |
|---|---|---|---|---|
| (Intercept) | 42.46 | 0.23 | 185.83 | 0.00 |
| gift | 12.48 | 0.32 | 38.84 | 0.00 |
Split and A/B Testing designs to derive inferences about customer behavior.
School supplies company wants to test optimal color scheme for new line of backpacks, geared toward high school market. Three schemes developed following qualitative research.
| Element | Execution |
|---|---|
| Independent variables | Age, Gender, Grade Level, Collage Image of Backpack Design |
| Dependent variables | Intent to Purchase score |
| Control/Experimental Groups | Participants randomly assigned to one of three groups, each group was only shown one backpack design scheme for evaluation |
| Randomization | Randomization: 750 US teenagers age 15-18 randomly sampled from an online sample provider |
Discussion:

| Situation | Online retailer wants to test the impact of a “value add” coupon after 2 purchases. |
| Hypothesis | Coupon will increase spending in subsequent purchases. |
| Method | Random sample of 1,000 customers. Half receive thank you note with coupon; half receive just a thank you note. Revenue measured 60 days after exposure. |
| Model Coefficients | ||||
| term | estimate | std.error | statistic | p.value |
|---|---|---|---|---|
| (Intercept) | 42.46 | 0.23 | 185.83 | 0.00 |
| gift | 12.48 | 0.32 | 38.84 | 0.00 |
| Model Fit | |||||
| r.squared | adj.r.squared | sigma | statistic | p.value | df |
|---|---|---|---|---|---|
| 0.60 | 0.60 | 5.08 | 1,508.80 | 0.00 | 1 |
| Situation | CPG company wishes to test a new campaign on Amazon that also includes new branding on packages. |
| Hypothesis | Both the campaign and new packaging will increase sales. |
| Method | Random sample of 1,000 customers. 2x2 design with each respondent seeing only one Amazon ad (A-Existing vs. B-New) and one package design (Y-Existing vs Z-New). |
| Model Coefficients | ||||
| term | estimate | std.error | statistic | p.value |
|---|---|---|---|---|
| (Intercept) | 17.67 | 0.03 | 653.02 | 0.00 |
| promoB | −1.88 | 0.03 | −60.26 | 0.00 |
| packageZ | 1.98 | 0.03 | 63.76 | 0.00 |
| Model Fit | |||||
| r.squared | adj.r.squared | sigma | statistic | p.value | df |
|---|---|---|---|---|---|
| 0.88 | 0.88 | 0.49 | 3,723.32 | 0.00 | 2 |
