Presentation on theme: "Research Methods in Unit 3 Psychology. A quick introduction."— Presentation transcript:
Research Methods in Unit 3 Psychology
A quick introduction
Hypothesis is ……….. Testable prediction of the relationship between 2 or more events or characteristics. It is usually based on knowledge of other research findings or theories on the topic being studied. Written statement Expressed clearly and precisely
Example Hypothesis “ This study is designed to assess the hypothesis that students with better study habits will suffer less test anxiety.” Unless your study is exploratory in nature, your hypothesis should always explain what you expect to happen during the course of your experiment or research.
SO In other words, the researcher is hypothesizing that the independent variable causes the dependent variable and is doing her experiment to test this hypothesis.
How do we start to research? The type of research method made by the researcher depends on which method is most appropriate for the specific topic of research interest. SimilaritiesDifferences All use rules and procedures Use of sample of participants who provide data for the research
2 types Experimental Research: IV DV Extraneous Variables Experimental Control groups Sampling procedures Can be manipulated and controlled Descriptive research: Case studies Observational studies Study aspects of behaviour and mental process as they occur
WE CAN RESEARCH SOMETHING BY USING EXPERIMENTAL RESEARCH WHERE
Experimental Research Experiment is used to test whether one variable or ‘thing’ influences a change or causes a change in another variable It is a collection of research designs which use manipulation and controlled testing to understand causal processes. Generally, one or more variables are manipulated to determine their effect on a dependent variable.research designs
Wait a minute WHAT do “conceptualization” and “operationalization” mean??? And WHAT is a “variable”?
Good question!!! Here are some definitions!!! A variable is the “thing” that you’re interested in studying—like depression or gender or levels of emotionality (how emotional someone is) or different types of food!
Variables Independent Variables (I) vary Varied or changed in some way or form “cause” Dependent Variables Outcome Depends on independent variable “effect” or end result Extraneous variables any other variable that could cause a change in experiment
Let’s try a question to make sure you understand these terms… Dr. Brain wrote the following question on the board: “Why do some students succeed academically (whereas others fail)? “ In this question, are we approaching academic success as an independent variable or a dependent variable? – In this question, academic success is a DEPENDENT variable because we’re trying to figure out what causes it. – If we believe that having clear goals causes some people to succeed in school whereas others fail, then we are interested in studying the presence or absence of clear goals as an independent variable, a possible CAUSE of academic success or failure.
To “operationalize” a variable is to decide how you will measure it For example, if the variable you’re interested in is depression: – Will you ask people to rate themselves, and if so, on what sort of a scale? – Alternatively, will you measure depression by facial expression? By some behavior that you observe? In some other way? For example, if the variable you’re interested in is depression: – Will you ask people to rate themselves, and if so, on what sort of a scale? – Alternatively, will you measure depression by facial expression? By some behavior that you observe? In some other way?
To “operationalize” a variable is to decide how you will measure it If the variable you’re studying is intelligence & you don’t think GAT test is a good measure of intelligence, what measure WILL you use? Asking these sorts of questions is completing the process of “operationalizing” your variables. By the way, conceptualization & operationalization are necessary for ALL the different research methods (not just for naturalistic observation)
The construction of actual, concrete measurement techniques; the creation of “operations” that will result in the desired measurements. The development or choice of specific research procedures (operations) that will result in representing the concepts of interest. The construction of actual, concrete measurement techniques; the creation of “operations” that will result in the desired measurements. The development or choice of specific research procedures (operations) that will result in representing the concepts of interest.
Operationalising What does it mean? Strictly define the variables We are trying to make something more measureable We operationalise hypothesis, IV and DV’s There are THREE steps in operationalization: a. Formulating Concepts into Variables b. Formulating Variables into Measures c. Formulate Instruments for the Measures Each of these steps is considered below
Operationalization also sets down exact definitions of each variable, increasing the quality of the results, and improving the robustness of the design. design
Try this example
What is wrong with this.... This question is a little bit fuzzy What does it mean children ? Grow which way? What does more quickly mean? 1 yr 10 yrs etc
Practicing Operationalisation When you need to operationalise something you need to include the following: a.the variables b. the identity criteria for each variable. c. a measurement procedure for each variable d. what would count as evidence for or against the hypothesis.
If we operationalise this hypothesis... Hypothesis Children grow more quickly if they eat vegetables.” Identify what age group What vegetables The amount of time the test will be taken over Does the sample of kids reflect the wider community Already we have operationalised this hypothesis
How does controlled experimental design eliminate or deal with extraneous variables? 1) First, it eliminates as many extraneous variables as it can by standardizing the experimental procedure so that all groups experience the same thing
For example: as we have discussed, placebo (such as “sugar pills”) are sometimes used to make sure the control & experimental group do not differ on the extraneous variable of “believing the treatment will work.” Placebos make sure that ALL groups have this same belief. Remember the goal is to make EVERYTHING the same between the experimental and the control group EXCEPT for the independent variable?
Some more terms…. There are some subjects who are administered the independent variable (in this case, study groups) and some subjects who aren’t The group of subjects to whom the independent variable is administered is called the experimental group. The other group is called the control group. The control group and the experimental group should be the same in all other ways. The only way in which they should differ is on the independent variable.
experimental group. The group in an experiment who is exposed to the independent variable control group. The group in experiment not exposed to the independent variable, used for comparison with the experimental group.
The choice of the sample is critical…. – The sample must be large enough in order for the researcher to be able to generalize to the population. (I shouldn’t interview two students and then say what all Bluffton students think!) – The sample also needs to be representative of the population, so for example, I shouldn’t just talk to seniors…. or to men… or to white students… or to religious students, etc. If I am interested in saying something about ALL Bluffton students, I need to talk to a sample of people that adequately represents all of the differences in the population.
M and M activities
Selecting Participants Represenatative Sample 1)Random sampling every member of population of research interest has equal chance of being selected 2)Stratified sampling dividing the population to be sampled into different sub groups or strata then selecting a sample from each group
Different subgroups Random stratified sampling a random selection from each sub group, get accurate lists of people within each stratum Random allocation also random assignment participants selected for the experiment are just as likely to be in the experimental group as the control group
DATA Qualitative - data involving qualities or characteristics of a participants experience of what is being experienced Quantitative – in development a change in the quantity or amount of thinking, behaving or feeling. Numerical value
WHAT is “correlational data”????? A correlation is a relationship between two variables. When two variables are correlated, that means they are related to each other in one of two ways: – Positive correlation: as one of the variables increases, so does the other – Negative correlation: as one of the variables increases, the other decreases A correlation is a relationship between two variables. When two variables are correlated, that means they are related to each other in one of two ways: – Positive correlation: as one of the variables increases, so does the other – Negative correlation: as one of the variables increases, the other decreases
Correlational Data – Two variables can be positively or negatively correlated or not correlated at all (unrelated) – Note that negative correlations indicate that there IS a relationship between the variables, the relationship is just an inverse one. Just shows as x goes up Y goes down
Correlation does NOT mean causation Correlation tells us nothing about the direction of the relationship between two variables or whether either of them really causes the other
Researchers use statistics to analyse and describe the data that they collect, they also use it to help them interpret the results obtained from the research
Inferential and Descriptive statistics Descriptive- used for analysing, organising, summarising and describing the results Inferential- used for interpreting and giving extra meaning to the results
Descriptive Mean ( average) – could be used to describe the average performance of a particular thing MEASURE OF CENTRAL TENDENCY- central or average value in a set of scores Median- middle score of mid point Mode- most frequently occurring score
Who cares, I here you say…. You would use Mean, Median and Mode to indicate trends in the population IF you need to comment on some results and need a single figure you could calculate the mean, median and or the mode and discuss
Inferential These statistics allow researches to draw conclusions based on evidence Allow the researcher to make conclusions and generalisations
Types of Inferential Statistics Statistical significance – is there a real difference between the control group and the experimental group, that is not due to chance
Interpreting P-Values Looking at the probability of something occurring Reliability Internal consistency Construct and external The lower the p-value the less probable of it occurring The higher the p-value the more probable of it occurring
P-VALUE – yay !!!! To test to see if results are by chance or not 5 or fewer times (<5) in 100 repetitions P<0.06 would indicate that there was a 6% chance ( 6 or less in100)that the difference in the scores was due to chance alone
STAND UP AND CLAP YOUR HANDS !!!
Some more examples P<0.01 (less than or equal to 1 in 100) P<0.001 (less than or equal to 1 in 1000
Order effect The effect of administering treatments in a specific order. Are you the first or second or last participant can that impact on the results The effect of administering treatments in a specific order. Are you the first or second or last participant can that impact on the results
Placebo effect The phenomenon in research where the subject’s beliefs about the outcome can significantly effect the outcome without any other intervention.
Single and double blind procedure Double blind procedure is a method of enhancing internal validity in an experiment. In double blind procedure, neither the researcher nor the subjects are made aware of which group is the experimental group and which the control group.experimental groupcontrol group 2 groups do not know what is going on(double)
Single Blind procedure A testing procedure in which the administrators do not tell the subjects if they are being given a test treatment or a control treatment in order to avoid bias in the results Subject does not know- Single (1)
Types of experimental research designs Repeated Measures- P’s act as their own control, increases sensitivity to detect differences, Matched participants- participant pairs are matched on variables relevant to the DV to reduce variability between groups, difficult and time consuming Independent groups- participants are randomly assigned to only one of the conditions to ensure rough equivalence between groups on many unknown and therefore, uncontrolled variables
Brain imaging techniques Allows us to see activity of brain SPECT is one example Single Photon Emission Computed Tomography (SPECT) allows us to visualize functional information about a patient's specific organ or body system.
DESCRIPTIVE RESEARCH Descriptive research: Case studies Observational studies Study aspects of behaviour and mental process as they occur
Case Study In depth study of an individual or small group of individuals Notice that the study is “in depth”!! Some case studies involve spending hours, days, months, years with a particular person to understand them thoroughly Used most often in study of rare phenomena, e.g., people with particular types of brain damage or other rare conditions, serial killers, particularly creative people or people with other rare abilities, etc.
Survey / Interview
Surveys/Interviews Questioning individuals through paper & pencil, phone interviews or face-to-face interviews May ask about just one variable or may gather information on multiple variables in an attempt to study the relationship between them
Who you choose to survey or interview is very important. For example, you may want to know what Bluffton University students prefer to eat in the cafeteria, but you will probably NOT ask each & every Bluffton student in finding this out. Instead, researchers survey or interview a smaller number of people (called a sample) who are expected to represent the entire group of people in which you are interested (the population). Researchers then generalize the answer they get from the sample to the population. For this to work, the choice of the sample is critical. For example, you may want to know what Bluffton University students prefer to eat in the cafeteria, but you will probably NOT ask each & every Bluffton student in finding this out. Instead, researchers survey or interview a smaller number of people (called a sample) who are expected to represent the entire group of people in which you are interested (the population). Researchers then generalize the answer they get from the sample to the population. For this to work, the choice of the sample is critical.
In Surveys/Interviews… The wording used for the questions is very important. A minor change in wording might result in very different answers. Who you choose to survey or interview is very important. – Researchers rarely interview ALL the people in which they are interested
In conclusion A conclusion is a decision or judgement about what the results obtained from an investigation mean.