Suppose that we are interested in understanding whether eating a balanced diet can reduce obesity. Let bal be a dummy variable denoting whether a person consumes a balanced diet, and weight be a variable denoting the person's weight in pounds weight = \"[30 + \"Blbal + \"[32educ + \"[33inc + \"B4gym + \"[35bal.gym Suppose you estimate that \"I32 = 3. Can you come up with an example of a standard error for the variable educ such that it is significant at a 10% level of significance, but not at a 5% level of significance? Can you come up with an example of a standard error such that it is now significant at 5% but not at 10% level of significance