Students typically come to an economics class with many misconceptions, not just random errors but systematic biases (see especially Caplan 2002).
Bill Goffe recently (2009) surveyed one of his macro principles classes and found, for example, that the median student believes that 35% of workers earn the minimum wage and a substantial fraction think that a majority of workers earn the minimum wage (Actual rate in 2007: 2.3% of hourly-paid workers and a smaller share of all workers earn the minimum wage, rates are probably somewhat higher today since the min. wage has risen and wages have not).
When asked about profits as a percentage of sales the median student guessed 30% (actual rate, closer to 4%).
When asked about the inflation rate over the last year (survey was in 2009) the median student guessed 11%. Actual rate: much closer to 0%. Note, how important such misconceptions could be to policy.
When asked by how much has income per person in the United States changed since 1950 (after adjusting for inflation) the median student said an increase of 25%. Actual rate an increase of about 248%, thus the median student was off by a factor of 10.
I would add that there are also theoretical misconceptions that are probably even more important than factual misconceptions. For example, I think it would be useful to ask questions such as:
1. A number of new furniture manufacturers open in North Carolina, since the new competition forced existing manufacturers to reduce prices the effect of this additional competition was to reduce wages for workers in the furniture industry.
Rate this argument on a 1 to 10 scale from least to most plausible or likely.
2. Competition from Korean automobile manufacturers caused US manufacturers to cut quality in order to reduce their costs. Rate this argument on a 1 to 10 scale from least to most plausible or likely.
3. A law that prevents supermarkets from advertising the prices of their products will reduce the supermarket costs and in turn this will reduce prices to consumers. Rate this argument on a 1 to 10 scale from least to most plausible or likely.
Note that the point here is not to say that an answer is wrong but to understand the implicit models that students are using. It can help a teacher to know what these conceptions and misconceptions are in advance so that they can be addressed.
To further this research, Bill Goffe is putting together a colloborative database on economic misconceptions that teachers are welcome to contribute towards. See also Bill's page on the large literature on teaching in physics much of which is also relevant to economics.