[David L.]

Abstract
Do opposites really attract? Or is it, “birds of a feather (omit comma) flock together?”

We began our study with our hypothesis that a positive correlation exists between couples with similar personalities and high relationship satisfaction levels.
Unless this is for Psychology Today or some such low-key publication that encourages a chatty, person-to-person style (as opposed to a scientific journal), do not personalize your writing with personal pronouns etc. Even if for just a presentation to other students, you should maintain the objective standpoint.)



This study investigates the hypothesis that a positive correlation exists between couples with similar personalities, and high relationship-satisfaction levels.
(Use of the comma separates the variables being used in the correlation. Note use of the hyphen to join relationship and satisfaction. Note also: you are investigating the hypothesis of a relationship between the variables, but the null hypothesis to be tested is of 'no relationship'.)



In order to determine a correlation between these two variables, thirty male-female couples were asked to complete two tests. The Myers-Briggs Type Indicator
(use the correct name of the test)


and the Couples Satisfaction Index by Funk and Rogge. (The (I think you mean 'then') both-omit)

Omit:
The male and female participant of each couple completed the tests individually from their partner. Both the male and female partner's results were recorded to examine their level of similarity and relationship satisfaction

Instead:
All participants completed the questionnaires individually; and results for partners collated to determine (examine their level-omit) degree of similarity, and relationship-satisfaction. Upon examination of all the test results-omit)

It was concluded that there was a significant positive[/color] correlation between couples with similar personalities, and high relationship-satisfaction levels. (See below)

I was surprised that only a medium-strong correlation exists between the variables.
Firstly, you should have determined whether the correlation itself is statistically significant ie greater than zero. A correlation of .33 might well still not be statistically significant depending on the sample size. Having so determined that it is significant (that is, a real correlation between the variables exists) then a correlation less than .33 is regarded as "weak" whilst .34 to .66 is "medium" and, and .67+ is "strong". Even in the astract, the actual correlation coefficient should be quoted, and significance level eg Spearman's rho = .69 (p<.05).

I had expected to find that couples that had a higher level of similarity would score significantly higher on the relationship satisfaction index because of the belief that couples that share more similarities are more likely to be successful.
(This is unnecessary, as it was the whole basis of the study)
However we have accepted our hypothesis that a positive medium-strong correlation exists between personality similarities and relationship satisfaction.
The way this is phrased is: your null hypothesis would have been that NO CORRELATION exists and a significant positive correlation would have then allowed you to reject this, and find support for the experimenial hypothesis, the adage, that indeed in personal relationships, like-feathered birds form a flock.
Did you think that ONE psychological test would assess ALL that is involved in what attracts couples? What about looks, financial standing and prospects, educational background, highly specific interests, religious/no religious affiliation? One test is not going to account for all the variance. You seem to be implying you expected something pretty close to +1. I reiterate: that a correlation was found, significantly different to zero, is the most that could have been determined experimentally.