How much do you "know" already?: are you planning a quantitative or qualitative study?

Do you need to do more exploratory work/ research? Or do you need to test already formulated hypotheses or propositions?

What exactly do you want to learn or establish?

Dear Dr. Brian Barnard

As you know, Yemen one of the developing countries, and the prevalence rate of CMV among general population is greater than 95%.

We conducted two studies to find out the prevalence of CMV infection among pregnant women and blood donors. The result was that the vast majority had CMV antibodies. Although, the proportion of Yemeni pregnant women with primary infection and women with seronegative CMV infection was negligible. However, the risk of the birth of a fetus with birth Defect is highly, especially in the absence of awareness about the virus and asocated risk factors among pregnant women.

Because there is no effective vaccine or treatment to treat CMV infection during pregnancy and a lack of knowledge about the virus among pregnant women, as demonstrated by our findings, there was a need to measure the extent of knowledge about CMV among health care workers.

Considering that the disease is most significant in pregnancy due to the associated congenital malformations and fetal deaths, questions should include knowledge about

- basic ante-natal care

- infections in pregnancy

- TORCH infection

- immunisation

- antenatal diagnosis of congenital abnormalities

- MTP

- CMV - have they heard of it, do they know the significance

- maternal and infant mortality rates

i can understand that, if there is no true treatment for the disease, knowledge as a type of prevention is the next best thing or step.

in a sense, approaching healthcare workers makes the assumption that healthcare workers are the link in the chain - the best medium, and the first/ best point of contact/ intervention. this may actually be a grounded assumption. you would know.

so, based on this, one could also take a step back and look at the factors that impact awareness of the disease among (pregnant) women, and in the society in general. from that, one can devise the best strategy, and confirm that healthcare workers will be the best approach or point of contact/ intervention.

personally, i would also be interested in how the disease is seen, both by women, the population, and healthcare workers.

this would be impacted by health and healthcare philosophy too.

so, you can consider either a quantitative or qualitative study, and a very scientific or medical study, or a very/ more social study.

you can both/ either use surveys with a strong medical or scientific base, or semi-structured interviews to get more personal views, and to really understand the context and situation.

Dear Dr. Ashwath Ramji

Thank you so much for your kind reply.

That's what I'm looking for but I'm also looking at the nature of the question formula in order to choose the appropriate statistical test for the analysis of the data.

Dear Dr. Brian Barnard

This is true. We believe that the gap here may be due to the lack of knowledge of the health workers about CMV and the absence of awareness about this virus among pregnant women in Yemen is due to the lack of education of women by the health workers about the virus and the risk factors associated with it

i would address any deficiency in knowledge also and equally with research on current thinking, views, beliefs, etc, to understand why the knowledge deficiency exists in the first place.

i suppose research into the knowledge of healthcare workers regarding the disease would reveal whether the knowledge gap (also) exists with healthcare workers.

at the same time, i revert back to social factors - how serious the disease is taken, etc. the social factors that permit the disease to persist.

you may not be able to resolve the matter with one research study only.

examples of interview questions that would reveal knowledge, thinking, and perceptions regarding the disease, would include:

- do you know someone with the disease?

- how has the disease affected you or your family?

- how dangerous or risky do you belief the disease to be?

- do you believe the disease can be prevented?

- what do you see as the remedy or cure for the disease?

- have you considered any steps to prevent the disease?

also from these questions, some of the propositions are that either the population/ women are not impacted extensively by the disease - the disease is not close enough to them to really take it seriously. a disease typically become common knowledge, if enough people are affected by it. or, that they dont believe the disease can be cured or prevented - part of the knowledge gap you highlight.

i am not sure you need any true statistical test when surveying healthcare workers. you would make sure you have a proper representation in your sampling, or substantiate a non-random sampling technique.

you would use descriptive statistics to analyze and unpack the results.

you can test across clusters or groups within the data. but a lot will only emerge after having obtained the data, i believe.

if the disease is attributed to lack of awareness or knowledge, what about alternative methods to raise awareness, other than healthcare workers? for instance, what about a national awareness campaign?

how well researched or documented is the prevention of the disease - is it a straightforward or complicated matter? this too would impact healthcare workers' knowledge.

Great inputs

RG FELLOWS

Following

Sir, if you want to be able to derive the data and analyse it directly from the answers on the questionnaire without an intermediate data analysis step then you should think along the the lines of assigning binary keys or limited alpha-numerical denominators to each answer/option.

WIth multiple-choice or ranking type of questions you would be able to derive quantitative results and with text fields qualitative results would be the straightforward derivatives. Simple or cross tabulation methods may also be employed here.

your statistical test is dependent on your research questions/ propositions or hypotheses.

surveys normally go with hypotheses or propositions.

if your hypotheses or propositions are not clearly formulated, which might be your case, you can go with descriptive stats, and switch to something more elaborate after data analysis, provided that you collect sufficient background information as well.

the descriptive stats will reveal your research propositions and hypotheses to you - you let the data talk to you.

Review literature.

Prepare a validated questionnaire

Go ahead

after a literature review, you may still be left with research questions rather than research hypotheses or propositions. that is perfectly possible and normal. it shows that not enough research has been done to date, particularly in specific contexts. then explorative research is in order. you would like take it step by step. perfectly fine to start with descriptive statistics