Dichotomous Data:

- Data that  fall  into only  two  categories are called dichotomous data. 

- Although these  data ordinarily might be considered nominal level  (e.g. pass-fail, qualified-unqualified), they also can be treated as  interval level in many statistical tests because the variable either has  an underlying continues characteristics such as pass-fail or is  conceptually changed  to represent the presence or absence of a characteristic. 

Levels  of data  can  be measured  in various ways,  e.g. multi-model measurement,  occurs when several measures instead of just  one are  selected, specially when measures  are  indirect or complicated, such  as growth, compliance.  In  these cases, indicators are used to measure the phenomena.  For example, height, weight, head circumference and  arm measure  can  be  used as indicators  of growth. 

The  best approach in planning research is  to  select, if possible, variables that  can be measured at interval and ratio level  so  that all options  are available  to  the researcher for statistical analysis.