ki 1 Raya Abu-Tawileh Dana Alrafaiah Hamza Alduraidi 1 P a g e
Today we are talking about introduction to biostatistics, its purpose, fields of studies found in this topic and philosophies of quantitative research and biostatistics. What is Biostatistics? It is the knowledge of summarizing big data into more organized and understandable pieces of information to better understand phenomena in the field of human health and social sciences. *We in medicine and other health care professions are very interested in understanding biostatistics to understand populations in term of health issues. So, biostatistics can also be defined as the application of mathematical and statistical probability laws, rules, regulations, and equations to understand populations. But in order to achieve that, first we have to understand the individual and the small group of individuals to understand the populations. The process of moving from sample level to population level is called inferences; which means building conclusions about a population using knowledge acquired from a smaller group of individuals (of the same population) known as the sample. When the sample is selected it has to be manageable; large enough to be representative of the population, but suitable in size to suit the available-often limited- resources. Biostatistics is concerned with: 1) collection 2) organizing 3) summarizing 4) analysing 2 P a g e
-Collection, organizing and summarizing make up the first part which is called descriptive statistics. we collect data from several individuals, but it's not feasible to try to understand the whole data while its raw. اي بشكل خام وعشوائ Ex: we want to know the level of happiness in a classroom on a scale of 1-10; 1 being least happy and 10 being most. You take a sample of 25 students(subjects) and ask them the question. You ll end up with 25 different numbers that aren t meaningful yet. Meaning it won t be necessarily helpful to say that person X answered the question with 5, person Y with 6 etc; these numbers do not indicate the level of happiness in the classroom as a whole. They re not helpful or manageable and most importantly cannot be used in the next step: analysis. So, we do descriptive statistics to describe these huge sums of data into more organized, summarized and understandable pieces of information, in the form of tables and numbers using mean, median, standard deviation, percentages and proportions. So, the data becomes more describable in the least amount of words and numbers possible. Back to the example: we find the summation of the 25 numbers and we divided them by the number in the sample(n), this will give us the mean or average which can be analysed to draw conclusions. -Analysing is the second part which is called inferential statistics. Analysing means comparison; we compare groups of people in terms of their sample statistics and after this we draw inferences to better understand the population. This way there s no need to give an observation to each individual alone to make conclusions about the whole population. 3 P a g e
-In every research there is a first step of collecting data, then a second step of entering this data to an appropriate software-usually SPSS(will be discussed later)-, then data will be cleaned, organized and summarised(managed) so it s easier to deal with, and then it is presented as tables, graphs or specific numbers that are ready to be analysed, and finally the results are taken and compared to other groups in order to come up with inference regarding the population, which is the most important step in any research. **** Biostatistics aims to make things easier to understand by summarizing them, making smaller pieces of information more meaningful and help researchers test their hypotheses. Hypothesis is an assumption; for example, we hypothesize that refugees living in a refugee camp have a lower quality of life compared to people who live outside these camps. This is just a hypothesis; it isn t tested or proved. And can come from different sources of knowledge, theory and information. In order to prove it, it has to be tested using biostatistics. How? We will recruit two samples; one will be of people who live in the camp (not everyone inside the camp will be recruited), and the second-with the same number of people- from outside the camp. Next, we collect data from both samples by asking what is the quality of life in scores for every individual in both sample. Then we find the mean for both samples and compare them to get the final numbers. If the final numbers were statistically significant احصائية معت ربة( )داللة this will tell us about the difference in the quality of life between inside and outside the camp populations, and since it is significant it can be generalised to the rest of the population. 4 P a g e
By this we can say we added knowledge regarding the difference in quality of life inside and outside the camp that was unknown before and was proved not by ideas in mind but by evidence. Data: Data are huge sums of information about individuals in the sample. However, data as it is collected cannot be used during the entire process of research as it s not meaningful enough; we need to convert it into information that are meaningful, standardized and useful in making the inferences. Resources of data recruiting a sample then reviewing the answers regarding the questionnaire item is a direct method of collecting data prospectively, other methods include: 1- previous records in hospitals, clinics and health care facilities. 2- surveys: how happy are you? (a simple one item survey). 3-counting: for example, if you re only interested in knowing the number of people diagnosed with a certain condition. 4- experiments: when you want objective answers that aren t altered by the individual; saliva sample to know the level of nicotine. 5- reports: can be from governmental agencies (you can refer to them when you don t have resources to collect new data yourself or when the sample is too big). The source of data depends on the nature of the research itself; its question, purpose, what are you trying to collect etc. After obtaining the data we convert it into meaningful numbers to draw conclusions and reach knowledge. 5 P a g e
Knowledge can be collected by: -previous experiences. -practice: in clinics or hospitals. -reading text books. - reading literature, which is all the published research in journals regarding certain topic. -from experts who have this knowledge. -Research, which is the most sophisticated form of obtaining knowledge. As evident, obtaining knowledge varies, and so the academic field divided into different schools of thoughts that regard different sources as the most important sources of knowledge, such sources include: -Philosophy. -Logical theorizing. -Theology (holy scriptures) -Empirical studies. Research: it is an attempt by scientists to understand the universe, to produce and bridge gaps in our knowledge. Since the beginning of time humans have tried to reach perfect understanding of the universe, not just the pure sciences like physics and chemistry but also human behaviour and attitude. It can be built on studies that are carried out by: 1- scientific method: it is the most prestigious way to gain knowledge. It is the process of gathering data from small numbers of people and trying to generalize it to a bigger picture, that leads to bridging the gap in humanity s knowledge and understanding of phenomena. Here we start from the smaller to the bigger; from understanding the individual to 6 P a g e
understanding the group of individuals then the population and then hopefully the universe. 2- logical reasoning: it's either deductive or inductive reasoning: a) deductive reasoning: its type of logic which starts from general into particular. اي من العموم اىل الخصوص او من الكل اىل اىل الجزء Starting from population and then conclude into individuals Example: All men are mortal(general) Socrates is a man. (particular) Therefore, Socrates is mortal. (conclusion) b) inductive :)االستقرائية( reasoning it involves going from a series of specific cases and observations then collecting all data from individuals and making conclusions regarding the whole population based on that data, it must have mean or median scores. Example: you have this matrix: 6,13,20,27 Can you inductively reason what comes after 27? Each time we are adding a 7 so 34 will come next. Inductive reasoning is about understanding the population by starting with individuals, and that is what scientists do in scientific methods and in quantitative research. ان تبدأ بالمعطيات المفرده ح ى ت تصل بالنهايه اىل االجوبه الشموليه العامه وهذا يلرت احدى prediction وه ال biostatisticsاهم احتياجات ال Inductive reasoning gives us the ability to understand and predict answers and so make modifications in the future. It is the best way of building knowledge and to come up with the concept of research. There are two types of research: qualitative and quantitative. 1) qualitative research: it's about getting very small sample without careful selection or criteria in sampling, then the researchers sit with 7 P a g e
each individual and ask them open-ended questions to explore their understanding of criteria concept, then highlight common words repeated by the individuals, these common words are called themes, it models how people tend to answer certain questions. 2) quantitative research: which is scientists' favourite type of research, it works more systematically and depends more on mathematics, probabilities and statistics. Paradigms: It either depends on 1)Ontology 2)Epistemology 3)Axiology 4)Methodology: Quantitative researches in health care sciences love to mostly rely on methodology(etiology), which starts with individuals and moves on to the population. Major paradigms: 1) Positivism: believe that reality is only one, true and exists, and all must be able to realize it, associated with deductive reasoning. 2) Naturalism: believe that reality is multiple, subjective and mentally constructed by individuals, associated with inductive reasoning. لكل انسان الreality خاصه به In naturalism human conception varies from one individual to another. Example: when looking at a wall, positivism claims that this wall is real and exists outside our personal perception. That it existed and will continue to exist as it is regardless of our ability to conceive it. Naturalism claims that since you use your eyes and brain to conceive its existence, and since everyone has different eyes and a different brain, it 8 P a g e
is not accurate to claim that we all see the wall in the same manner, and that this uniqueness should be acknowledged, respected and taken into consideration while we obtain information in positivism people do quantitative research in which we consider each individual in the sample to have the same perception (because everyone shares the same set of tools that are anatomically and functionally similar) and then their answers should be standardized, and we should take into consideration that every one of these individuals understands the questions equally, then we come up with the mean score which can be easily generalized using biostatistics. اي انهم يؤمنوا ان الب ر ش معياريي while in naturalism they don t generalize; they believe that every individual has his own unique perception. (in naturalism they ask each individual separately) To conclude, in positivism they believe that people can be standardized, a sample of certain people could be representative of the whole population and certain close end questions are good enough to be asked to everybody equally and at the same time to collect data from them, then come up with a conclusion regarding the population. While in naturalism every individual has his own way of understanding, that s why we can't generalize unless we ask every single individual separately. according to doctor's opinion the best is to mix between the two paradigms by using quantitative and qualitative researches. -quantitative recruits a large sample from a population, the individuals here are known as subjects. -qualitative recruits a smaller sample randomly (we don t care how representative it is as we don t want to generalise) and interviews each one individually and allows them to express their experiences, individuals here are known as participants. But for logistic reasons it's not always feasible and realistic to do mix methods studies to answer certain research questions, but in our 9 P a g e
doctor's point of view he would always tend to do quantitative research because it's more valid, although qualitative might be much credible; because every individual has the chance to express their experience. to conclude quantitative and qualitative complete and complement each other. It's important to know that we can't say we are 100% sure and confident we understand certain phenomena; in quantitative research we can say we are 95% confident and it is acceptable in scientific society. Statistics don t allow us to understand 100%. All questions that are in your mind about differences between quantitative and qualitative researches will be discussed in the coming lectures. This was an introduction on how to understand researches Best wishes Raya Abu-Tawileh 10 P a g e