correlations and meta analysis

A correlation is an analysis of the relationship between co-variables. Co-variables consist of two variables that are measured by the researcher and then compared to each other, e.g., age and memory ability.

True. In correlational research, the co-variables are not manipulated. There is no independent variable; instead, the two co-variables are measured and compared with one another.

c. Skewed correlation is not a type of correlation.

A negative correlation is where one co-variable increases and the other co-variable decreases  (but not necessarily at the same rate).

A scattergraph is used to present the results of correlational analysis.

The correlation coefficient represents both the direction and the strength of the relationship between the co-variables, expressed as a value between -1 and +1.

b. A correlation coefficient of +0.3 shows a weak positive correlation.

Correlations use large amounts of quantitative data, making this method of analysis high in reliability.

c. A limitation of **correlations is that they only work well for linear relationships such as height and shoe size. They are less successful **when dealing with non-linear relationships, e.g., the number of cups of caffeine in a week and IQ (there is no correlation between these covariables).

One difference between correlations and experiments is that correlations cannot show cause-effect, they can only show a relationship between co-variables.

Bi-directional refers to the idea that each co-variable may influence the other without any clear indication as to which co-variable affected the other. E.g., exam scores may be linked to days of absence from school, but the absence could be due to exam stress or to other factors entirely.

Unlike correlations, an experiment measures the difference in participant performance depending on manipulation of the independent variable.

A meta-analysis is a quantitative research method that takes data from published studies and analyses it to find the effect size for a specific variable.

False. A meta-analysis is an example of secondary data. Other researchers have conducted the research and published the results. E.g., a meta-analysis of 133 studies to investigate cultural variations in conformity.

b. The results of a meta-analysis are expressed in terms of effect size, which is a score from 0 to 1 that indicates the strength of the effect.

True. There is less chance of bias confounding the results of a meta-analysis due to the use of secondary data. The researchers have not carried out the research, so they cannot have influenced the outcome in any way.

The use of secondary data in a meta-analysis limits its reliability. The researchers cannot be 100% confident as to the degree of precision and control used by the original researchers.