"While conducting a social research project, chances are good that one will encounter indices and scales. These terms are often interchangeable. While these two types of measures have some characteristics in common, they are the two different concepts and have distinguishing features." (Babbie, 2001). Taking into consideration the project "The outcome of a child's future regarding the employment status of the mother", in my opinion, the Likert Scale is the best type of composite measure. This scale contains such categories, as: "strongly agree", "agree", "disagree" and "strongly disagree". The researcher can suggest questions, like: "Do you agree that children grow up "better" if their mothers work outside of home?" ; or create a statement - "It is better when the mother doesn't work and stays home with children". Then "a score of 0 to 4 is assigned. For example, a score of 0 means "strongly disagree", a 1 means "disagree", a 2 means "agree" etc." (Babbie, 2001). The scores for each question will then be summed up for each respondent to create an overall score.
I absolutely share the first student's opinion about indices and scales. According to his point of view, the last are "primarily used to predict effects..., indices are designed mainly for exploring" and as such steps as "examination, scoring the index and validating it" are used, it is obvious that the "index method seems to be more relevant".
I also agree with the second student's opinion that it is better to "create a true Likert Scale (mentioned above), calculating the average index score of the individual statements" of former criminals. Such debriefing cannot be easy and suggest only positive and negative critical answer. Creating several-leveled scale "tends to reach the upper limits of reliability." (Nunnally, 1978). This can lead to the better understanding of intensity among the indicators variable".