The results would have changed significantly - they'd be inaccurate. You'd be forcing someone who actually HAS a neutral belief to report that they have either a positive or negative belief.
There's a science to how many choices to provide as well as whether the midpoint is a "neutral/zero" or some other value. There is also a huge body of evidence establishing that you should use an ODD number of choices - either 5 or 7 (I won't get into that debate.)
Any Likert scale - used to measure beliefs or attitudes - will have either a unipolar or bipolar construct
Unipolar constructs lend themselves to an amount; either there is the maximum amount of the answer, the minimum amount of the answer, or something in between. However, since we're dealing with an amount the midpoint is not "neutral."
How helpful is this answer?
- Extremely helpful
- Very helpful
- Somewhat helpful
- Slightly helpful
- Not at all helpful
As you can see, this post can't be "Neutral helpful"
Bipolar constructs measure attitudes that fall on either side of midpoint that represents true ambivalence or neutrality.
How do you feel about taking market research surveys
- I love taking market research surveys
- I like taking market research surveys
- Neutral
- I dislike taking market research surveys
- I hate taking market research surveys
Forcing someone who genuinely holds a neutral feeling/opinion to choose a positive or negative position is problematic in two ways:
- People who hold a neutral position that are forced to choose will make their positive or negative choice randomly, so you haven't really "forced" anything
- You will deprive yourself of the ability to learn that people are neutral about something, which can often be important to know