Hypothesis Flashcards
This is my hypothesis. What would be a reasonable null hypothesis?
Anxiety and pain are strongly correlated.
There is no relationship between anxiety and pain
A research question can be constructed using initialisms like PICO, which stands for Population, Intervention, Comparison or Control,
and Outcome. Using the following information, construct a reasonable research question.
P = teenage mothers
I = financial incentive to attend school
C = no intervention, standard care
O=completion of GCSE or A-levels
Are teenager mothers who are offered a financial incentive to attend school more likely to complete GCSEs or A-levels
A research question can be constructed using an acronym such as SPIDER which stands for Sample, Phenomenon of Interest, Design,
Evaluation, and Research Type. Using the following information create a reasonable research question.
S=young parents
PI=antenatal education
D=survey
E=experiences
R=mixed methods
What are the experience of young parents who attend ante-natal education classes? A mixed methods study using focus groups and surveys.
Give two reasons why we prefer a randomly selected sample from the population in quantitative research? Provide a brief explanation
for each.
A sample that is selected at random from the population will generate a less biased result than a convenience sample. This is because convenience
samples are drawn from easily accessible groups, and groups of people are more likely to share certain characteristics. For example, taking a
sample from a cohort or university students will bias the group in terms of level of education.
A sample that is drawn at random from a population are likely to be heterogeneous - this means that it is more likely to represent the diversity of the population (Careful because this answer is quite similar to the first - so a bit more is needed to differentiate it). A diverse sample will make the results of the research more generalisable.
Phase 1 of my study is to understand the reasons for late
arrival in class. What sort of sampling strategy should I use
and why?
There is more than one correct answer here.
A purposive sample of those who are late will give you a better
understanding of the issues students face that make them late.
However, you could also do a convenience sample including some
who are late and some who are not. This might help you gather intelligence
about what helps or motivates students to be on time as well as what makes
them late.
In a class of students you could also use stratified random sampling where you take a random
sample of those who are late, and a random sample of those who are not. This would work well
if you have large numbers of students and you are doing a survey study - but it is probably not the best
choice for understanding something from the students’ perspective. A survey can close down your answer set
too much.
Phase 2 of my study is to identify the prevalence of each of
the factors identified in Phase 1. In other words - how many
students experience the issues that cause lateness? What
sort of sampling should I employ in this phase to collect a
sample of 100 from a student population of 1000.
Random sampling
My hypothesis is that lateness to class has an association
with module marks. What sort of hypothesis is this?
Two-tailed
One-tailed
Zero-tailed
Cannot tell from the information given.
Two-tailed
I undertake a Mann-Whitney U test to determine whether
the median module mark is different for those who are late
more than 5 times, and those who are late less than 5 times
in a module. The p value returned with this result is p=0.65.
What does this mean?
There is a significant difference between the two
groups but we cannot tell from this information which
group did better.
There is a significant difference between the two
groups and the group who were late less often did
better.
There is no significant difference between the two
groups but the group who were late less often had
higher median marks.
There is no significant difference between the two
groups and while there may be a difference between
the medians we do not know who what it is.
There is no significant difference between the two
groups and while there may be a difference between
the medians we do not know who what it is.
