Graphs should communicate information clearly and accurately. A title gives the context of what the graph is showing, the axis labels identify what is being measured on each axis and often include the units, and the data must be plotted accurately so the graph truly represents the values or trends in the data. When any of these elements are missing or wrong, the graph can be misinterpreted or leave readers unsure about what they’re looking at. For example, without a title and labeled axes, a reader wouldn’t know what comparison or time period the graph represents, and plotting data inaccurately can distort conclusions. A properly constructed graph, with a descriptive title, clear axis labels, and correctly plotted data, is easy to read and interpret, making it the best choice.

The key idea is that quasi-experiments study cause-and-effect without random assignment of participants to groups. In these designs, the independent variable isn’t allocated by the researcher in a random way, often because it’s a pre-existing or naturally occurring condition rather than something the researcher can assign. That’s why the description that the independent variable has not been determined by anyone and the variables simply exist fits best: it captures the lack of random assignment or full experimental control over how groups are formed. If the IV were randomly assigned, you’d be looking at a true experiment. The requirement for ethical approval is separate and not what defines a quasi-experiment, and an IV that is clearly manipulated by the researcher with randomization missing is still characterized by the lack of random assignment rather than by the existence of the variable itself.

The mode is the value that occurs most often, so finding it is basically a matter of counting how frequently each value appears and picking the highest frequency. That straightforward counting makes the mode very easy to calculate, especially in small data sets or when you have a clear frequency table. It’s also particularly useful for categorical data (nominal data) because you can simply see which category occurs most often. Keep in mind, though, that the mode isn’t always the best summary of a data set. It can be unstable if there are several values with the same highest frequency (multi-modal distributions), and it doesn’t tell you anything about how the rest of the data are spread or how large or small the values are. The other statements are misleading: the mode does identify the most common value, and it can be useful for categorical data, so those points aren’t accurate.

Repeated measures design uses the same participants across all conditions, so individual differences between participants don’t add as much uncontrolled noise to the data. That means the manipulation’s effect is easier to detect because the error variance is reduced. With this greater sensitivity, you can achieve reliable results with fewer participants, which is why the advantage is described as needing fewer participants. Note that order effects can still occur, so researchers usually counterbalance the order of conditions to control for that, rather than assuming randomness alone fixes the issue. Also, while individual differences between people still exist, they are held constant across conditions, not eliminated, and the claim that there are no individual differences isn’t accurate. The idea that it is more time consuming isn’t an inherent advantage; in fact, this design can be more efficient in participant numbers even if it requires more management of order effects.