¶ … Measurement and Concepts of Validity "As you move up the ladder of measurement, the amount of information that is gained increases. At the lowest level, you have only categorization" (Measurement, n.d.: 21). The lowest level of measurement is the nominal scale which "simply places people, events, perceptions, etc. into categories based on some common trait" (Garger 2010). For example, a group of people deemed to be overweight in a medical study would be a nominal category. With every successive level of measurement "you add knowledge about the order of the categories included" (Measurement, n.d.: 21). Thus, with an ordinal scale, data is ranked from lowest to highest which makes the scale more information-rich than the nominal scale (Garger 2010). To use the above-cited example, this might include rankings of persons according to their weight in kilos. The next level above the ordinal scale, the interval scale classifies, orders, and also "define[s] how much categories differ one from another" (Measurement, n.d.: 21). For example, the amount of difference between each overweight person in the group would be defined. Finally, a "ratio scale requires all of these characteristics as well as a non-arbitrary, or true, zero value" (Measurement, n.d.:...

Because it cannot be assumed that "that the lowest possible score in your data collection automatically represents an absolute zero point" only with the knowledge conveyed by the ratio scale is the data-collector's knowledge 'complete' (Garger 2010). For example, if collecting data on obese people who scored from 1-10 on an test with a ratio scale which measured their risks for developing medical complications from obesity, it would be impossible to convert from an normal, ordinal, or interval scale to this ratio scale because there would be no information at the lower levels what the absolute zero point was but it would be possible to convert to the less information-rich systems of measurement from the ratio scale.
"The first question involves what those who study research methods call validity. Validity addresses the question of whether the variable used actually reflects the concept or theory you seek to examine" while in contrast the measure of reliability "addresses the question of whether a measure gains information in a consistent manner" (Measurement, n.d.: 22-23). A valid research design is a design which accurately measures the phenomenon it is supposed to measure: for example, if a design was to test the validity of a low-calorie diet…