is it advisable to create a mixed effects model for these data? I have accuracy values for two conditions. My data are quantitative & continuous and range from 0-1 (worst to best performance). They come from two independent conditions with a single value per participant. Originally, I compared the distribution of values for each condition to zero (one-sample t-test), and then I compared values between conditions (two-sample t-test). 
A reviewer thinks that I can create a mixed model to account for both of these steps, but I'm unsure of its implementation (or if it's even possible).  If it is possible/sensible, I'd like to establish (1) whether the data from each condition are statistically significant from zero and (2) whether they differ significantly between conditions using a mixed model. My null hypotheses would be (1) the distribution of accuracy values is not significantly different from zero and (2) there is no difference in accuracy values between conditions.
 A: Taking into consideration your follow-up comments:


*

*If there are lots of zeros (i.e., a floor effect), the dependent variable will not be normally distributed. You may want to consider instead a logistic regression where you categorize responses as 0 or greater than 0. This depends on other theoretical considerations, but it seems to me like that's what you might be trying to do?

*You describe a between-subjects study (participants in condition A or B) where each participant is measured 240 times. You could fit a multilevel model that looks like:
DV ~ Condition + (1|ID)
Which means that the DV is predicted by Condition and a random intercept (i.e., intercepts are different for each participant).
There are two levels in this dataset:
Level 1 is the observation. Each participant's DV is measured 240 times at this level.
Level 2 is the participant. Each participant has a value for the Condition (0 or 1).
You say you want to get a different slope for each person, but that would mean you need some predictor at Level 1, because the slope for condition can't change between people... because people aren't measured in both conditions.
