RELIABILITY TEST refers to the reproduction of the result of prior research. This is to contrast with VALIDITY TEST which is the test for precision, i.e. how does the result of the experiment coincide with the expected value.
Sometimes the term "reliability" is used to mean "trustworthiness" of the data set in a sense of "data integrity." In such a case, the term data integrity is more accurate so that there would not be any confusion with "reliability test"----in statistics.
DATA INTEGRITY is the test of the integrity, i.e. trust worthiness of the data, i.e. primary or secondary. The integrity of both data type may be tested by looking at the following factors:
-Collection method
-Storage method
-Data treatment
-Data control
-Data accessibility
-Completeness, accuracy & consistency
We might look at other relevant factors as well, i.e. objectivity in data collection (not collecting data to serve some interested [selective collection]). If the data has integrity, one can say that it is trustworthy; therefore, any analysis and conclusion drawn from such a data set may meet scientific or research investigation.
REFERENCES: Some articles are attached for data integrity.