One of the key problems with writing a hypothesis is to make sure that your variables are clearly operatoinalized - that is, that when someone reads your study, the variables are clearly defined.  This helps to increase the validity of the study - that is, it guarantees that you have measured what you claim to measure.

The research hypothesis predicts whether the IV will lead to an increase or decrease in the DV.  It is the null hypothesis that predicts that there will be no significant effect of the IV on the DV.

The dependent variable - that is, the variable that is measured - is "memory."  However, this is too vague to really be an acceptable DV for a research hypothesis; it must be more carefully operationalized.  Therefore, the best answer is one that gives a specific definition of how memory will be measured - such as the number of words remembered from a list or the number of images remembered after one week.

Social desirability effect is when a participant acts in a way that is "socially acceptable" in order to not appear abnormal, rude, ungrateful, prejudiced or otherwise socially unacceptable.  This means that their responses or behaviours are not really what they think or usually do, but it protects their public image and self-esteem.  This is commonly seen in interviews as well as in overt (not covert) observations.

This study simply compares the text messaging spelling with the scores on a spelling test.  No variable is manipulated to see how it may affect their spelling. 

All of the other answers are true of an independent samples design.  In a repeated measures design you only have one group that receives both treatments; therefore, you don't have to worry that one group was "better at math" or "better memorizers" than the other group, as you have the same participants in both conditions.

The independent variable is what the researcher changes (manipulates) in order to measure the effect on a dependent variable.  In this case, the type of therapy is what is different for each group of participants.

Because correlational studies take place "post-facto" - that is, after the fact - we have no idea whether or not the news coverage had an effect on the individual. We cannot know, unless we ask all fatal car crash victims, whether they had seen the coverage.  This is a logical fallacy often relevant to correlational researcher - just because something happened before something else, it does not mean that it caused it.  Post hoc ergo propter hoc.

The independent variable is the presence of a smoking ban, but the smoking ban is not manipulated by the researcher; the smoking ban is imposed by the Czech state.  The study is then a repeated measures where a community's level of smoking is measured before the ban and then after the ban.  This is a natural experiment.

Order effects will most likely affect this study.  Participants may get better at memorizing lists as the experiment continues.  This is called the practice effect.  They may also just get bored or tired of the experiment, thus decreasing performance.