Reliability Refers to the Consistency

¶ … Reliability refers to the consistency or repeatability of the measurement, I .e. that it can be extended to many other situations, whereas validity refers to the accuracy of the measurement, whether it accurately asses the results that it purports to evaluate and whether it is an appropriate match.

As per ethical issue, voluntary informed consent is important when conducting experiments that involve human participants. Voluntary informed consent refers to the fact that the participant is given an entire understanding of the study in a manner in which he or she can comprehend and is then given the liberty to truly abstain or to drop out of the study whenever she wishes

Longitudinal research is research that extends over a long breadth of time (usually over several years); cross-sectional research is research that spans a varied population / various regions. The advantages of the first is that results can be tested and re-tested; history effects (and maturational), however, occur where the study becomes 'old' to participants, and time changes certain aspects, so researcher may receive different effects over time. Cross-sectional reaps a diversity of experiences from a heterogeneous population; on the other hand, it is far more difficult to match subjects to each other therefore confounding effects may more easily creep into study.

4. (i) Criterion validity -- comparing the results with another measure e.g. comparing the results of IQ tests with school tests. Advantages: that one can gain greater objectivity. Disadvantages: each measure tests slightly different elements, and each was taken at a different time when confounding elements may have interfered with results. (ii) Construct validity -- investigating whether the measure is really measuring the theoretical validity it is supposed to be. For instance, intelligence may be defined according to many different constructs. A measure would have to take all into consideration in order to correctly measure it. The advantages are that the measure is as thorough -- and, therefore, as accurate as possible. The disadvantages are that there are, occasionally, numerous conflicting and, occasionally, oppositional ideas about attributes of the construct (e.g. intelligence). These constructs, too, may change with time. In that manner, a measure may be constantly questionably valid.

5. To measure reliability, we can use two procedures: (i). Test-retest i.e. we measure more than once. The advantage: there is reinforcement to result. The disadvantage, transformations may be introduced the second time e.g. maturational effect, history and so forth, leading to a different result. Another problem is cueing where subjects might adjust to the test itself. (ii) the Alternative Forms Procedure: This involves two tests given two or more times. The advantages include less cueing because the tests differ. Disadvantages include history and maturation factors.

6. Positive approach is empirically based and insists that measures must proceed according to strict evidence and their validity must be generated from experimental (or quantitative) research. The advantages are that this accords greater validity and reliability. The disadvantages are that the complexity of human factors is inadequately addressed by the rigidity and inflexibility of the experimental method. The qualitative (or non-positivistic) on the other hand, insists that humans cannot be measured or tested in a laboratory environment. The results are, however, less reliable, and oftentimes more subjective.

7. Cross sectional research is research that is conducted over different population groups at a single space in time. For instance, assessing whether people of different cultures vary in the way that they walk. Longitudinal research is conducted on one population over a period of time for instance to see whether this style of walking changes over the years and if so in which way.

8. Social workers do have an ethical responsibility to engage in research since information changes constantly, and if social workers wish to be as professional and as efficient as possible they need to keep up with current research. For that reason, caregiving should be evidence-based rather than transmitted from hearsays.

9. Characteristics of a normal bell shaped curve include the fact that is has a continuous distribution; is symmetrical about the mean; each side of the mean can go on forever in either direction; the curve is unimodal (the values are bunched up against the center of the curve) the area under the curve is 1, so each side being symmetrical, the area on each side of the mean is 0.5.