Week 17 the scientific method lecture lm
Download as PPT, PDF1 like573 views
This document provides an overview of the scientific method and importance of research skills. It discusses how human knowledge is often flawed and unscientific, developing through guesses and speculation rather than scientific evidence. It then covers the scientific method, which involves generating hypotheses based on initial observations and theories, then collecting data to test the theories. The document outlines the key stages of the scientific method and importance of validity, reliability and analyzing data to make inferences beyond the sample. Finally, it discusses discussing the implications of findings and how they can inform future research directions.
Week 17 the scientific method lecture lm
- 1. Dissertation Preparation (SPO039-2) – Lecture 2 Notes by Lee Taylor The University of Bedfordshire Department of Sport and Exercise Sciences Contact: lee.taylor@hull.ac.uk 1
- 2. Why are Research Skills important as Sports Scientists / Coaches / Personal Trainers? • Human knowledge and beliefs are often flawed, unscientific and irrational in their development eg: - ~50% of the population of the USA do not believe in evolution by natural selection! - 47% believe in UFOs! - 49% believe that it is possible for someone to be physically possessed by the devil! - 26% believe in mental telepathy!
- 3. The basis of knowledge •The basis of knowledge in the real world can be unsubstantiated guesses, hunches, or speculations! —Scientists are not exempt from these personal beliefs, prejudices, faith, hope, hunches and myths. Therefore •To develop new knowledge, as scientists, we must follow guidelines/rules specific to developing new knowledge, aka The Scientific Method! —The primary aim in following The Scientific Method is to avoid these biases in research – scientists should be as objective as possible.
- 4. Take home message of why research methods are important! • As sport scientists, personal trainers and coaches we need to ensure the theory we use in our practice, whether that be as an educator, coach or trainer, is correct. — To ensure this is achieved any evidence we use should have been proved / collected / analysed using the guidelines mentioned previously, i.e. The Scientific Method, to reduce biases and anecdotal practises.
- 5. Revisit to Scientific Method What are the two general approaches to research? 1.The scientific method (hypothetico- deductive method) 2. The unscientific approach. 5
- 6. Unscientific methods are reliant on: Intuition Commonsense and gut reaction Authority Teacher, parent – reliability? Tenacity Beliefs, superstitions, no logical foundation Data Identify Variables Measure Variables • Graph Data • Fit a model Initial Observation (Research Question) Generate Theory Generate Hypotheses Collect Data to Test Theory Analyse Data Discuss the meaning and implications of the findings The scientific method Unscientific Method Could be argued that that all are necessary to some degree to answer specific research questions The best method? 6
- 7. Scientific approach Systematic: Identification and labelling of variables and the design of a test/method to test any relationship between variables. Logical: Examination of the procedures used in the research process allows researchers to evaluate the conclusions drawn. Empirical: Researcher collects data on which to base decisions. Reductive: Researcher uses individual events (data) to establish relationships. Replicable: Research process recorded to enable others to test the findings (validity and reliability) and use previous results as a basis for future Unscientific approach Tenacity: An unscientific method of problem solving in which people cling to certain beliefs regardless of the lack of supporting evidence. Intuition: Common sense or self evident, however, many self-evident truths are subsequently found to be false, e.g. the Earth is flat. Authority: Reference to some authority, e.g. a teacher / religious leader / renowned academic as a source ok knowledge – their knowledge can be bias/wrong Galileo vs Ptolemy. The rationalistic method: Knowledge derived by reasoning can be flawed by simple premises e.g. Basketball players are tall. Tom Thumb is a basketball player. 7
- 8. Data Identify Variables Measure Variables • Graph Data • Fit a model 1) Initial Observation (Research Question) 2) Generate Theory 3) Generate Hypotheses 4) Collect Data to Test Theory 5) Analyse Data 6) Discuss the meaning and implications of the findings The scientific method 6 key stages of the hypothetico-deductive method 8
- 9. RQ, Theory and Hypotheses • Research Question: Is a broad question related to a topic of interest which tentatively indentifies that there may be a relationship among variables - e.g. Hypoxic training and sea level athletic performance. • General Theory: A hypothesised general principle or set of principles that explain known findings about a topic and from which new hypotheses can be generated - e.g. hypoxia effects VO2 max. • Hypotheses (experimental hypothesis): Are specific predictions, underpinned by previous observations and knowledge which are the “best guess”, “educated guess” or “expected result” - e.g. VO2 max decreases in hypoxia. 9
- 10. 1. Initial Observation (research question) • In a nut shell “come up with a question that needs answering”: Mini Task 1 Using your provisional dissertation topic, write down your broad research question(s). - Explain to the person next to you what you are generally going to investigate. 10
- 11. 1. Initial Observation (research question) • In a nut shell “come up with a question that needs answering”: • SES? • SSPT • SSC - Does altitude training improve sea level performance? - Does bicarbonate loading improve repeated sprint performance? - Do super sets increase muscle mass (hypertrophy/hyperplasia)? - Is static or dynamic core strength training key to reduced lower back pain? - Does whole body vibration exercise maintain muscle temperature at half time? - Are double sessions twice as effective compared to single sessions in pre-season training? 11
- 12. 1. Initial Observation (research question) • Once you have a broad research question, this needs to be refined: e.g. why is Usain Bolt the fastest man on the planet? - Many factors, or variables need to be considered to answer this question: 12
- 13. 1. Initial Observation (research question) • Once you have a broad research question, this needs to be refined: for your research questions what factors could be considered? - Many factors, or variables need to be considered to answer this question: 13
- 14. 1. Initial Observation (research question) • Once you have a broad research question, this needs to be refined: e.g. why is Usain Bolt the fastest man on the planet? - Many factors, or variables need to be considered to answer this question: Genes Training Attitude Environment Biomechanics Nutrition It is the considerations of these factors or variables that allow more specific theories to be generated. 14
- 15. 1/2. Initial Observation and theory generation Genes: A predisposition for explosive power and adaptation to training, fibre type distribution, phenotype to avoid injury. Training: An effective and appropriate regime. Attitude: Positive, willingness to train, accept criticism. Environment: Weather (conducive to training) / Coaches/ physios / sponsors Biomechanics: Long Limbs, perfect gait. Nutrition: Adequate, well balanced nutrition From further consideration and greater exploration of these concepts we can now generate specific research questions/theories. 15
- 16. 1/2. Initial Observation and theory generation 16
- 17. 1/2/3. Initial observation, theory and hypotheses 1. Calvin has an initial observation (1) and research question (2). 2. 1 and 2 ultimately help form his hypotheses (3). 11 2 3 17
- 18. Data Identify Variables Measure Variables • Graph Data • Fit a model 1) Initial Observation (Research Question) 2) Generate Theory 3) Generate Hypotheses 4) Collect Data to Test Theory 5) Analyse Data 6) Discuss the meaning and implications of the findings The scientific method Lets work through stages 1 – 3 with an example 18
- 19. 1/2/3. Initial observation, theory and hypotheses Before we can generate an hypothesis (3) using our RQ (1) and GT (2), we must identify the independent and dependant variables. Independent Variable (IV): What the researcher is manipulating. Dependant Variable (DV): Is the effect of the independent variable. Example: Research question: The effect of hypoxia (altitude) on an individuals VO2max? Theory: Hypoxia does effect VO2max. Hypotheses: Hypoxia will decrease VO2max. Independent Variable = Hypoxia / normoxia Dependent Variable = VO2max 19
- 20. 1/2/3. Initial observation, theory and hypotheses Mini Task 2 Using your provisional dissertation topic refine your research question(s), general theory and specific experimental hypothesis(es), ensuring you identify your variables. Example: Independent Variable (IV): What the researcher is manipulating. Dependant Variable (DV): Is the effect of the independent variable. Research question: The effect of hypoxia (altitude) on an individuals VO2max? Theory: Hypoxia does effect VO2max. Hypotheses: Hypoxia will decrease VO2max. Independent Variable = Hypoxia / normoxia Dependent Variable = VO2max 20
- 21. 3. Hypotheses generation – guidelines Remember when generating hypotheses the following processes should have been adhered to: 1) The hypotheses is normally generated by indentifying a research question and then a suitable theory. 2) This process and the resulting hypotheses should be based on a theoretical construct and/or results of a previous study (scientific underpinning – the reading of articles!!!!!!!!!!!). 3) The researcher should identify both the independent and dependant variables. 21
- 22. Data Identify Variables Measure Variables • Graph Data • Fit a model 1) Initial Observation (Research Question) 2) Generate Theory 3) Generate Hypotheses 4) Collect Data to Test Theory 5) Analyse Data 6) Discuss the meaning and implications of the findings The scientific method 22 We have now conducted the first three phases of the scientific method
- 23. 4. Collect data to test theory. So in line with our hypothesis that Hypoxia will decrease VO2max: We need to measure the variable(s) we have identified: Independent variable: Hypoxia / Normoxia (environmental chamber) Dependant variable: VO2max (treadmill, direct or method) The data collection process is often considered the easiest part of the research process – often routine, such as repeated VO2max tests. However......... Planning the method is one of the most difficult steps of the research process A good method will help you by: - Maximising the validity and reliability of the study - Ensure the data collected is not biased, i.e. is representative of the population 23
- 24. Good results need Validity and Reliability Good methods maximise internal and external validity, and reliability: Validity: Evidence that a study allows correct inferences about the question it was aimed to answer, or, that a test measures what it set out to measure conceptually. Internal validity: The extent to which the results of the study can be attributed to the treatments (interventions) used in the study. “Alternative explanations” - subject performed extra training outside of intervention, they took a supplement....hence why experimental control so important control...control...control External validity: The generalizability of the results of the study. People, places, times, random sample selection, etc. Use control groups and randomisation to augment external validity. Duplicate samples also augment external validity. Reliability: Whether an instrument can be interpreted consistently across different situations, the results repeated, can include equipment / protocols. Significant results are not “one of events” and are inherently repeatable, e.g. other scientists should be able to perform exactly the same experiment and generate the same24
- 25. VO2max 50 ± 5.0 ml.kg.min-1 Pre VO2max 52 ± 3.0 ml.kg.min-1 Post VO2max 70 ± 5.0 ml.kg.min-1 10 wks - Live High Train Low Altitude Training -3SD -2SD -1SD 50 ml +1SD +2SD +3SD Frequencyofascore 68% 95% 99% Population Mean = 50 ± 5.2 -15ml -10ml -5ml 50ml +5ml +10ml +15ml 35ml 40ml 45ml 50ml 55ml 60ml 65ml Internal Validity
- 26. VO2max 50 ± 5.0 ml.kg.min-1 Pre VO2max 52 ± 3.0 ml.kg.min-1 Post VO2max 70 ± 5.0 ml.kg.min-1 10 wks - Live High Train Low Altitude Training 35ml 40ml 45ml 50ml 55ml 60ml 65ml 55ml 60ml 65ml 70ml 75ml 80ml 85ml Post VO2maxPre VO2max
- 27. 6. Generalisations can only be made about your sample. 5. Descriptive Statistics Central Tendency & Dispersion 4.Data collection 5. Inferential statistics Comparison of means, regression and correlation. 6. Generalisations (inference) can be extrapolated from your sample data to the wider population your sample is drawn from. 5. Data tested and analysed Sampling 27 External Validity
- 28. Data Identify Variables Measure Variables • Graph Data • Fit a model 1) Initial Observation (Research Question) 2) Generate Theory 3) Generate Hypotheses 4) Collect Data to Test Theory 5) Analyse Data 6) Discuss the meaning and implications of the findings/results The scientific method We have now conducted the first four phases of the scientific method 28
- 29. 5. Analysing Data Statistical analysis is performed to aid acceptance or rejection of the null hypothesis. Two types of statistics are generally employed: Descriptive statistics: Allow you to describe the basic features, i.e. the central tendency (mean, median & mode) and variability/dispersion (standard deviation & range) of your data. They can be used to consolidate a large amount of data into a simple summary. They do not allow you to make judgements beyond your sample to the population your sample was taken from. Inferential statistics: With inferential statistics, you are trying to reach conclusions that extend beyond the immediate data alone. Specifically, we use inferential statistics to make judgements of the probability (P value) that an observed difference or relationship within or between groups is a dependable one or one that might have happened by chance. This allows us to draw conclusions that extend beyond the immediate data generated from your specific sample, therefore, allowing inferences to made to population your sample was taken from. 29
- 30. 6. Generalisations can only be made about your sample. 5. Descriptive Statistics Central Tendency & Dispersion Data tested and analysed Sampling 4.Data collection 5. Inferential statistics Comparison of means, regression and correlation. 6. Generalisations can be extrapolated from your sample data to the wider population your sample is drawn from. 5. Data tested and analysed Sampling 30
- 31. Analysis Parametric Non-parametric Relationship between two variables (continuous data) Pearson correlation Spearman signed rank test Relationship between two variables (categorical data) -------- Chi-squared test for no association Comparing 2 means (different participants) Unpaired (independent) t test Mann-Whitney U test Comparing 2 means (same participants) Paired (dependent) t test Wilcoxon signed rank test Comparing 3 or more means (different participants) One-Way between-subject ANOVA Kruskal-Wallis test Predicting a dependent variable from one independent variable Simple linear regression -------- Predicting a dependent variable from more than one predictor variable Multiple linear regression -------- Comparing 3 or more means (same participants) Repeated measures one-way ANOVA Friedman test Comparing 2 or more means (different participants) but two between-subject variables Factorial between-subjects ANOVA -------- Comparing 2 or more means but one within-subject variable and one between-subject variable Mixed model ANOVA -------- 31
- 32. 6. Generalisations can only be made about your sample. 5. Descriptive Statistics Central Tendency & Dispersion Data tested and analysed Sampling 4.Data collection 5. Inferential statistics Comparison of means, regression and correlation. 6. Generalisations can be extrapolated from your sample data to the wider population your sample is drawn from. 5. Data tested and analysed Sampling 32
- 33. Data Identify Variables Measure Variables • Graph Data • Fit a model 1) Initial Observation (Research Question) 2) Generate Theory 3) Generate Hypotheses 4) Collect Data to Test Theory 5) Analyse Data 6) Discuss the meaning and implications of the findings/results The scientific method We have now conducted the first 5 phases of the scientific method 33
- 34. 6. Discussing the meanings, implications , future directions and real world applications of findings. Example: Our hypotheses that Hypoxia will decrease VO2max was proved correct: -To gain an accurate measure of an individuals true VO2max, tests must be conducted at sea level. - Comparison of VO2max values obtained at altitude (hypoxia) can not be securely made to those obtained at sea level. - Training intensity is lower at altitude, compared to sea level, and thus altitude training reduces training load – hence why the modern paradigm is “live high train low”. 34
- 35. 6. Generalisations can only be made about your sample. 5. Descriptive Statistics Central Tendency & Dispersion Data tested and analysed Sampling 4.Data collection 5. Inferential statistics Comparison of means, regression and correlation. 6. Generalisations can be extrapolated from your sample data to the wider population your sample is drawn from. 5. Data tested and analysed Sampling 35
- 36. Data Identify Variables Measure Variables • Graph Data • Fit a model 1) Initial Observation (Research Question) 2) Generate Theory 3) Generate Hypotheses 4) Collect Data to Test Theory 5) Analyse Data 6) Discuss the meaning and implications of the findings/results The scientific method We have now conducted the all phases on the scientific method 36