The research conducted by Sharon follows a non-experimental or correlational design. Correlation is a measure of the similarity, or relatedness, between two phenomena. When properly normalized, the correlation measure is a real number between —1 and +1, where a correlation value of +1 indicates that the two phenomena are identical, a correlation value of —1 means that they are diametrically opposite, and a correlation value of 0 means that they are uncorrelated, that is, that they agree exactly as much as they disagree (Goodwin 2009 325). As with any research design, correlational designs have both strengths and weaknesses, and it is important to be aware of these advantages and limitations when considering the results of research using them.
2. Ecological validity of Research
Sharon's research has ecological validity in the sense that her correlational research design, like designs in observational research, has identified the link between the happiness states of infants and their mothers' behavior in relation to post natal depression. In addition to this, Sharon data has come from real-world settings, that is, children playing in the crèche area are observed whilst their parents attend adult literacy classes. The infants mother are then required to fill up a questionnaire with the objective of measuring the degree of post natal depression. This high level of ecological validity assures us that the relation that we have found can be observed in everyday behavior in comparison to experimental research designs in which the researcher frequently creates relatively artificial situations in a laboratory setting.
Moreover, all correlational researches, including this, are helpful in deducing predictions. When two or more variables are correlated, we can use our knowledge of a person's score on one of the variables to predict his or her likely score on another variable (Weiten 2009 35-40). For example, if we know that intelligence and leadership skills are positively correlated, and then we can make a prediction those persons with leadership skills ought to possess good level of intelligence. Similarly, if we know a person's college grade point average, as well as his or her achievement test scores, we can predict his or her likely performance in graduate school. These predictions will not be perfect. But they will allow us to make a better guesses than we would have been able to if we had not known the person's score on the predictor variable ahead of time (Bondy et al 2010 43-52).
Despite their advantages, correlational designs also have a very important limitation. This limitation is that they cannot be used to draw conclusions about the causal relation among the variables that have been measured. An observed correlation between two variables does not necessarily indicate that either one of the variables caused the other. Let us imagine for a moment that a study has found a positive correlation between group cohesion and group performance and consider several possibilities about why these two variables might be correlated. Although one possibility is that increases in group cohesion ...