Hypotheses are educated guesses. A hypothesis results when the questions are transformed into statements that express the relationships between variables like an "ifâ€¦then" statement. According to Salkind (2009), a good hypothesis should: be stated in declarative form, not as a question; posit an expected relationship between variables; reflect a theory and literature upon which they are based; be brief and to the point; and be testable.
A hypothesis that may be constructed for research question 1 is: "The topics of the visual coverage of the Iraq War was reported in The New York Times more clearly and more fully than in The Guardian newspapers". This hypothesis is good because (1) the statement is forceful and in declarative form. (2) It describers a relationship between The New York Times and The Guardian newspapers on reporting of the topics of the visual coverage of the Iraq War. (3) Knowledge or literature about The New York Times and The Guardian newspapers should be useful in proving or disproving this hypothesis. (4) This hypothesis is brief and succinct. (5) This hypothesis might be tested. For example we can test this hypothesis by collecting all reports of the topics in The New York Times and in The Guardian newspapers for a period of time.
A hypothesis for research question 2 maybe following: "The New York Times reported the visual coverage of the human toll and destruction during the Iraq War more than The Guardian newspapers".
A hypothesis for research question 3 is: "In terms of nationality the visual coverage of the human toll and destruction during the Iraq War was reported in The New York Times more than in The Guardian newspapers".
In short, a good hypothesis meets each of five criteria above. A good hypothesis provides a transition from a problem statement into a form that is more amenable to testing using the research methods.
As Salkind (2009) reported "a good hypothesis should be stated in declarative form" (p. 29). It is good that this hypothesis is started in declarative form, not as a question. However the phrase "The hypothesis of this study is that" is not needed and should be removed. We are talking about hypothesis and this "introduction" to the hypothesis is not necessary.
The second criteria for good hypotheses is "it should posit an expected relationship between variables". We have in this hypothesis two variables as "lives" and "fitness centers" and relationship between them is "lives will be saved when fitness centers are required to have AEDs on the premises". So the following explanation "because when they are not legally required to have them they often do not purchases them" is not needed.
This hypothesis that given with some unnecessary phrases is long and complicated and it cannot be brief and to the point as a good hypothesis should be. And finally, the hypothesis is difficult to test because It does not describe clearly what should be tested. However we may find some literature upon which this hypothesis are based.
In short, the hypothesis is not so good due to lack of some criteria for "good hypotheses". The five criteria are effective tool to review and classify hypotheses.
On the article "To stream or not to stream in a quantitative business course" by Buhagiar & Potter the authors provided some literature cites. First they provided a definition of a term "video streaming" as "video streaming refers to the process of viewing video over Internet" by Boster (2006) or more detail: "Video streaming is a transmission of moving images over the internet in compressed form as a continuous stream. A recipient equipped with suitable 'player' software can decompress and view the images in real time." by Fill (2006). All literatures cited are recent, in last 5 years.
After that the article's authors mentioned about streaming video as a cost effective method to reach students and it has advantages over other media. "Videos are created easily and at low cost." by Clark (2007). "Streamed video is also different from CD -based or DVD - based video. Individual CDs or DVDs need to be produced and distributed to each user; not so for the streamed video." by Sheppard (2003). "Most
students appear enthusiastic about the use of video streaming technology." by Dupagne (2007). Therefore Literature review shows that streaming video is low-cost methods, convenience and liked by students.
The article's authors also mentioned some similar results of previous researches as:" Student performance in a distance course is similar to the performance of students in a face to face course" by Gagne (2001) or "there is no statistical difference between
face-to-face students and distance learning students" by Neuhauser (2002). Therefore the review of previous research in this article is related, recent and relatively complete.
Article's readers can understand the statement of the problem because the purpose of the study is clearly stated in the abstract. The purpose is tied to the literature that is reviewed. However the objectives of study should be more clearly stated in the introduction. Article's authors provided a previous research result about no difference on student performance of distance (web-based) and face to face courses. But "streaming video" is another method to deliver distance education. So the authors showed a reason of why the study is grounded and important to do.
The research hypothesis of article is an null hypothesis "There is no statistical difference on overall grades between a group of students learning in a quantitative
business course through video streaming with the option of going to a face-to-face lecture and a group of students learning same course only through face-to-face lecture." The research hypothesis is clearly and explicitly stated. The two variables in this article are "video streaming method with the option face-to-face" and "only face-to-face method". The research hypothesis stated a no relationship between these two variables. Indeed, the hypothesis state a clear association between variables.
Moreover, the hypothesis grounded in a review and presentation of relevant literature that is provided in the article's introduction. Of course the research hypothesis can be tested.
The dependent and independent variables in the article are clearly stated. two independent variables are two sections of ECO 3401: section 0004 with traditional face-to-face lecture, section 0L01 with video streaming. The independent variables are also number of students in each sections, number of exams and duration of studying. The dependents variables are grades of students of each group.
The study is clearly conducted and analyzed. Article's authors conducted 38 question tests using a level of significance of 0.05 and only two did not have a statistically identical mean: Question 28 and question 15. Face to face performed better on one and video streaming performed better on the other. The overall final exam average was for face to face 81.4% and for video streaming 81.9%, with no statistical difference (Î± = .05). However, the average age of the two groups was statistically significant: an average age of 20.9 for the face-to-face students and an average age of 20 for the video streaming students. While the difference between the average ages of the two groups is statistically significant, we don't believe this difference had a profound effect upon the overall outcome of our study.
This study uses convenience sampling. It is an example of nonprobability sampling. In this case the probability of selecting a single individual is depend on the national group that it belongs. The number of individuals from Germany, Italy and Switzerland are not equal and each group of individuals used different methods of recruitment. Participants in Germany and Italy were recruited via newspaper advertisements in only two city (Berlin and Milan), on other hand participants in Switzerland were recruited nationwide via newspapers and an internet site.
In this approach individuals are selected from three countries with different numbers of each country and different proportion of men and women (the German group has 228 women for 54 men, while Swiss group has 81 women for 158 men). So it is difficult to research something especially for men and women. We should not state about number of men and women in each group because it does give us any useful information for study.
The advantages of this approach are that it is low-cost and convenient. We can select any individual that accepted to participate on research. We don't have to pay attention to their gender. It is convenient for researchers. Moreover we can use different methods of recruitment for selecting individuals. It may reduce the cost for getting sampling. However this approach has disadvantages. With convenience sampling the degree of generalizability may be low. Because of difference of number of individuals on each group we cannot state anything about a result with nationality. It is the same problem with gender.
In short, this approach is convenience sampling. It is convenient and low-cost but it also has low degree of generalizability.
According to Salkind (2009), the level of measurement is the scale that represents a hierarchy of precision on which a variable might be assessed. The level of measurement used reflects how an outcome is measured. There are four levels of measurement: Nominal, ordinal, interval, and ratio.
Nominal: Attributes are only named. They might be literals or numerical values. Nominal level variables are categorical in nature. For example, name of subjects (geometry, literature, science,â€¦); type of computer (desktop, laptop, netbookâ€¦); group of students (group 1, group 2, group 3â€¦). There are some features of nominal level. First, the categories are mutual exclusive. A variable should belong to only one category. Second, if numbers are used as value, they mean simple a classification.
Ordinal: Attributes can be ordered. In this case distances between attributes does not have any meaning. Ordinal level variables often refers rankings of various outcomes. For example, ranking of colleges, ranking of students in a class,â€¦. The interval between values is not interpretable in an ordinal measure.
Interval: Distances between attributes do have meaning. Interval describes variables that have equal intervals between them. For example, the number of correct answers on a test, the temperature, Alice is 3 inches taller than Bobâ€¦..
Ratio: Absolute zero. Ratio describes not only variables that have equal intervals between them but also have an absolute zero. For example: height, weight,..
The levels of measurement is important because knowing the level of measurement helps us decide how to interpret the data from that variable. Moreover, knowing the level of measurement helps us decide what statistical analysis is appropriate on the values that were assigned. The more precise the level of measurement, the more accurate the measurement process will be.
The questionnaire has some problems. First of all, all the questions have to be numbered. Numbers would make the use of questionnaire information become easier. On the critique report we will number the questions of the questionnaire by numbers 1, 2, 3, and 4.
The first question has no meaning if a respondent is not involved in the decision of whish nursing home to choose. In this case, he or she wont continue answer to next questions from 2 to 4 and the contributed information is too small.
The question 2 is clear. Respondents can answer easily to this question.
The question 3 is not clear. Respondents have to answer to this question only if they answer "yes" to question 2. The question "did you use the Medicare Nursing Home Compare website?" is not easy to understand because respondents cannot clearly define meaning of question: using this website to find information about a nursing home or using this website for another purposes.
The question 4 is also not clear. It is difficult to answer to the question "did the website influence your decision in choosing a nursing home?". Cannot definitely answer "yes" or "no" to this question because making a decision usually be affected by many factors.
In short, the questionnaire is not useful and not helpful to you in collecting necessary data you want. It could not help you to accomplish your goal. The information could be obtained through another source.
A multiple-choice question includes the following components: the stem, which has the purpose of setting the question or posing the problem, the set of four alternatives (options). One of these options must be the correct answer, the other three should act as distracters.
An example of a multiple choice question:
10. What is characteristic of the web site:
B. Shows very fast
D. Presents many types of information
The stem of a multiple choice question "What is characteristic of the web site:" should be written as clearly as possible to reduce method errors.
There are four alternatives A, B, C and D. The right option is "A. Hypertext ". Three other option (B. Shows very fast, C. Hyperthreading, and D. Presents many types of information) are wrong and are called distracters.
A good distracter should be attractive enough that a person who does not know the correct answer might find it plausible. In this example, the wrong option "C. Hyperthreading" might be easily confused with the right option" A. Hypertext".
A good multiple-choice item has to discriminates between those who know the information on the test and those who do not. So an item that every body can answer correctly is not good. Otherwise an item that every body cannot answer correctly is also not good. These items do not tell the examiner who knows material and who does not. So we can maximize the discrimination by including the moderately difficult items.