Validity is whether or not you are measuring what you are supposed to be measuring, and reliability is whether or not your results are consistent.
If an instrument or experiment is valid, it will usually also be reliable as long as it is carefully constructed to control all variables except the one being studied.
“For a test to be valid, or truthful, it must first be reliable.” (educational assessment)
For example, if you are measuring the reading levels of students after a new computer-based reading program has been used, you will want to make sure that your test will get the same results every time it is taken, and it should if it accurately measures the students’ reading ability.
If an instrument or experiment is reliable, it does not necessarily have to be valid. You might get consistent results, but not actually be measuring what you think you are measuring. For example, if the above test of reading level was really a vocabulary test, you might get the same results each time but the results do not show reading level.