Between-subject and within-subject effects usually go together. For instance in a completely randomized design, the levels of the treatment factor are to be compared using the between-treatment sum of squares and mean square. The within-treatment effect is the error effect which is tested using the error sum of squares and the the error mean square.

Generally, I would recommend using a within-subject design for an experimental manipulation and counter-balancing the order of your conditions.  The two key advantages are: (1) greater power with fewer participants to detect an effect and (2) the ability to test additional hypotheses by correlating participants performance across condition.  However, if they apply, there are good reasons to use between-subject designs instead.  In particular, it's sometimes impossible to have a within-subject design because the experience of participants in one condition would distort responses in the other.  For example, if you wanted to see how people respond to the same vignette if the main character is male or female, participants would obviously recognize that they're reading the same story with different names; they would respond based on what they think the experimenter wants rather than the experimental manipulation.  Another reason to use a between-subject design would be if it would take too long for one participant to be in both conditions (i.e., fatigue).  On your first question, I agree with Ette that you can certainly have both between- and within-subject variables within the same study.  The statistics are the same, just different variations (e.g., Repeated-Measure ANOVA).  Best wishes with your research, Wissal. ~ Kevin

Hi,

you might want to read this publication

Charness, Gary, Uri Gneezy, and Michael A. Kuhn. "Experimental methods: Between-subject and within-subject design." Journal of Economic Behavior & Organization 81.1 (2012): 1-8.

No, the two concepts are not the same.