Applied Research Questions.
a-Descriptive and inferential statistics:
1-Similarities: Both are used in researches to make judgments on social behaviors, to find out the meanings of social incidents, to estimate the values of some materials or recreation activities. For example, a social worker likes to know the effects of a new program performed by another social worker on his or her patients. He or she needs statistics to have an exact picture of all the works that other social workers perform.
A football fan wants to know the trends of a team to make a bet on that team. An economist tries to understand why people nowadays like to spend less than ten years ago. All the “wants”, “likes”, and “needs” are the causes of a research that needs statistics.
a) Descriptive statistics are simple tools to describe the basic data. They give simple explanations about the samples and the performance. They also provide the simple quantitative analysis of the data. With descriptive statistics, people can read the data in a simple way. If there is a big amount of data, descriptive statistics will make it smaller by reducing them into small amounts and make them easy for readers to read.
b) Inferential statistics are more complex. They look into something beyond the simple data that they get. The inferential statistics are use to “reach conclusion that extend beyond the immediate data alone.” They use inferential statistics “to make judgments of the probability that an observed difference between groups is a dependable one or one that might have happened by chance in this study.” (http:www.socialresearchmethod.net/kb/statdesc.php
a-Single case and small-N research designs.
1) Similarities: Single case and Small-N research designs are observing cases. Single case research is idiographic rather than nomothetic. Small-N research have some characteristics but they are still idiographic, too.
2) Differences: Single case designs are often used in clinical psychology and neutro- psychology to know the meaning of some social activities. Even though they are single case, they are a combine of clinical observation, self reports, archive data. Single case research are represented by two types: case study and single-subject experimental designs. Also, single cases report the result of a treatment, a plan, or a new organization in a simple one. Single cases have limited validity. Single case research is the Extreme case of small-n research. Small-n research is more simple.
True experiment design includes more than one purposively created group, common measured outcome(s), and random assignment. “Sex and ethnicity do not satisfy this requirement since they cannot be purposively manipulated in this way” . True experimental designs occur when the sample are randomly assigned to program and comparison groups. If the experiments can be performed by randomly assignments, the experiment program can be considered as true designs. There are some threats “when the control group can be inadvertently exposed to the program: such a threat also occurs when key aspects of the program also exit in the comparison group.”
True experiment are “limited by narrow range of evaluation purposes they address.”
True experiment differs with experimental designs on the way the population, the ethnicity, and sex are designed. The threats to internal validity occur when the researcher tries to influence the outcome, that means, the researcher has a bias mind and change the dependent variables to have results as he or she wishes.
Quasi-experimental designs are commonly employed in the evaluation of educational program when random assignment is not possible or practical.” (http://pareonline.net/getvn.asp?v=5&n=14) Quasi experimental designs are non-equivalent, and posttest only.
For example, a researcher wants to find out the differences between two groups of students who will study in different topics. One group will study grammar, the other
study writing. At the end of program, students of both groups will have to pass a posttest to see which group studies better than other. There is a threat of this design
is that there is no way to compare or find out in details the real differences between
two groups, who are better in preparation, another group may be better already before