  Chapter 3: Misconceptions When people read, hear, or prepare research summaries, they sometimes have misconceptions about what is or isn't "sound practice" regarding the collection, analysis, and interpretation of data. Here are some of these common (and dangerous) misconceptions associated with the content of Chapter 3. A correlation coefficient does a better job of summarizing the strength and direction of a relationship between two variables than does a scatter diagram. If the correlation between the scores on two variables is very high, then the two means must be very similar. A correlation of .80 indicates twice the "relationship strength" as compared to a correlation of .40. A correlation never speaks to the notion of "cause and effect." If a single outlier is removed from a very large group, the value of r cannot change very much. An r of -.90 signifies a "low" relationship. If the correlation between two variables is equal to +.50 for a subgroup of men, and if the correlation between these same two variables is +.50 for a subgroup of women, then the correlation between these two variables will be +.50 for the combined group of men and women. There are commonly agreed-upon guidelines that clarify for researchers when they should use terms such as "strong," "moderate," and "weak" to describe relationship strength. A linear relationship between two variables exists only if the dots in a scatter diagram all fall on a straight line. If the researcher's data correspond to two variables that are qualitative in nature, it's impossible to compute a correlation coefficient.