5) You are the project manager of a large household survey using face-to-face interviewing methods to measure poverty levels in your country. You have carefully constructed an interpenetrated design assigning interviewers to ran- dom subsets of sample households within the PSUS in which they work. On average, each interviewer completes 25 interviews. From this design you have computed pmnt (according to the model reviewed in this chapter) that reflects interviewer variance. The following table provides a few of the results: Estimate Pint Percentage of households in poverty 0.09 Perteentage of households with unemployed adults 0.02 The project team receives the bad news that the next year's budget may be cut. You are asked to assess the impact on the precision of estimates if the number of interviewers is cut in half but the sample size is kept constant. a) If that were the only change made in the design, by what factor would the variance of each of the two estimates above change? b) What additional factors would you consider in addressing the proposed change in interviewer numbers?