One-way ANOVA: Etch Rate versus RF Power Source DF 55 MS F P RF Power 3 66871 22290 66.000 Error 16 5339 334 Total 19 72210 5 = 18.27 RSq = 92.61%...

In many integrated circuit manufacturing steps, wafers are completely coated with a layer of material such as silicon dioxide or a metal. The unwanted material is then selectively removed by etching through a mask thereby creating circuit patterns, electrical interconnections, and areas in which diffusions or metal depositions are to be made.  A plasma etching process is widely used for this operation, particularly in small geometry applications. Energy is supplied by a radio-frequency (RF) generator and causes plasma to be generated in the gap between electrodes. The chemical species in the plasma are determined by the particular gasses used. Fluorocarbons are often used in plasma etching

 An engineer is interested in investigating the relationship between the RF power setting and the etch rate for a particular tool. The objective of an experiment is to learn about the relationship between the etch rate and the RF power rate and to specify the power setting that will give a desired target etch rate. The experimenter is interested in a particular gas (C2F6 ) and gap (.80 cm), and wants to test four levels of RF power: 160 W, 180 W, 200 W, and 220 W.

2b) The following is the computer output from a one way Analysis of Variance. Evaluate whether mean etch rates are different for the different RF power values by testing the hypothesis that the means for the different power levels are the sameChoose the best answer from the following:

i) From the computer output the mean for 160 W is 551.20 which is different from the mean of 587.40 for 180 W, which is different from the mean of 625.40 for 200 W, which is different from the mean of 707.00 for 220 W. Thus, it can be concluded that the means for the different power levels are not the same.

ii) The R-Sq value for this model id 92.61%. This is very high, thus the data fits the model very well. This implies that the means are not all the same

iii) The p-value for testing the null hypothesis that the means are all the same is less than .05, consequently the null hypothesis is rejected. The conclusion is that not all the means are the same.

iv) The 95% confidence intervals do not overlap. This is strong evidence that the means are not all the same.

Choices:

i) From the computer output the mean for 160 W is 551.20 which is different from the mean of 587.40 for 180 W, which is different from the mean of 625.40 for 200 W, which is different from the mean of 707.00 for 220 W. Thus, it can be concluded that the means for the different power levels are not the same.

ii) The R-Sq value for this model id 92.61%. This is very high, thus the data fits the model very well. This implies that the means are not all the same

iii) The p-value for testing the null hypothesis that the means are all the same is less than .05, consequently the null hypothesis is rejected. The conclusion is that not all the means are the same.

iv) The 95% confidence intervals do not overlap. This is strong evidence that the means are not all the same.