It might sound strange for many newbies but match types do not affect QS (Quality Score). The AdWords system only collects QS data when the user's search query is an exact match of your keyword. If you use the phrase match keyword "video game" and the user keyes in *purchase video game* you may get an impression but these data will not be a factor in the calculation of QS for the keyword "video game". Since data is collected only for exact matches the QS of an exact match keyword is then assigned to the phrase match and broad match type keywords as well.
Now let's assume that only the broad match or phrase match type of a particular keyword is used in an ad group which has been running for quite a while to accrue sufficient data. If the broad match has accrued 100,000 impressions and 5,000 clicks , the CTR is 5%. In this case, you may experience that some of the users keying in a matching search term in fact do key in your exact keyword, say 10,000 out of the 100,000 users. Because in this case the numbers seem to be large enough chances are that approximately 500 of this 10,000 "tight" users will click on your ad resulting in 5% again. Data Matching This example shows that the assumption is false according to which you could enhance the CTR of a broad match keyword without achieving an increase in QS that is purely based on exact match CTR. Though QS data is not collected for a broad match keyword it *is* collected for the "embedded" exact matches.
Starting small is golden so newbie's performance numbers are usually not big enough. I mean say hundreds of or a few thousand impressions a week, and much fewer clicks. If you have ever tried to ride a bicycle you know that while it can be pretty easy at a reasonable speed, as you slow your vehicle balancing might get more and more difficult. A newbie may be in a similar situation if he wants to investigate the relationship between the CTR of a broad or phrase match keyword and its QS.
Paradoxically, he may experience a considerable increase in CTR associated with a worsening Quality Score or vica versa. (As set out above, if the performance numbers are really large, this scenario is practically out of question.) It may be frustrating and resulting in a distorted picture. It's perhaps even more lamentable that you get the least information in this respect in the most sensitive first stage of your learning curve. For this reason I'd advise every newbie to list their broad or phrase match keywords as at least 2 match types -- broad & exact or phrase & exact. That way at least they are likely to grasp more quickly what is happening and why to the relationship between the CTR of their broad or phrase match keyword and its Quality Score.