This method also has some disadvantages: - It cannot confidently establish the cause and effect relationship because so many other factors may influence the dependent variable. A lack of control reduces internal validity. This may also be a lack of a suitable control group. It is not easy to replicate such studies. It can only be used when conditions vary naturally. Such conditions are not always possible to find. Participants might be aware of being studied and show improvements just because of this. The independent variable is not directly manipulated; participants are not allocated at random to conditions.
Ethical conditions: - A natural experiment may involve withholding treatment from one group when a new educational programme is being tested and the research requires that one group of students do not have the new teaching method. If participants are unaware they are taking part, there is the issue of informed consent, researchers need to be sensitive to the problems of participants in unfavourable circumstances that may surround the behaviour being studied as in the case of deprived children.
There are different types of investigative methods that are not experimental, and these include naturalistic observation studies, studies using correlational analysis and interviews and questionnaire surveys
Naturalistic observation is when behaviour is observed in the natural environment. All variables are free to alter and interference is kept to a minimum. No independent variable is manipulated. The advantages of this method are: - They have high ecological validity, no problems of demand characteristics or evaluation apprehension, a richness of behaviour can be observed, gives a more realistic picture of spontaneous behaviour as the people tend to behave naturally, if the observer remains undetected, the method avoids most experimental effects, such as experimental bias.
The disadvantages of this method are: - It is not possible to infer cause and effect, it is difficult to replicate and therefore you can’t be certain that the result was not a one off, it is not possible to control extraneous variables, observer bias, the observer sees what he/she wants to see, when participants know that they are being watched i.e. disclosed observations they may behave unnaturally.
Ethical considerations: - Unclosed observations preclude the right to informed consent; disclosed observations may affect the individuals observed creating distress.
Correlation is a statistical technique used to quantify the strength relationship between two variables. Through correlational analysis one could investigate whether there is an association or relationship between two variables, for example whether there is a relationship between aggression and the amount of violent television watched. In correlational study no causal relationship can be determined. In this method a numerical value is calculated to represent the degree to which two sets of data are correlated. In terms of IV and DV they are not used because one does not depend on the other. The variable are called co-variables. Perfect positive correlation is +1.00, a perfect negative correlation is –1.00. If both variable increase together, this is called a positive correlation, if one variable increases as the other decreases this is then called a negative correlation.
The advantages of this method are:- Can be used where experimental manipulation would be unethical or impossible, it indicates possible relationships by co-variables and it might suggest future research ideas which would look at possible causal relationships. Two variables are correlated and you can predict one from the other. It is a useful method when manipulation of variables is impossible. Establishing no relationship between variables is useful in eliminating cause and effect. It is possible to obtain large amounts of data on a number of variables more rapidly and efficiently with experimental designs.
The disadvantages of this method are:- It does not establish cause and effect relationship, interpretation of the results is difficult, direction of causality is uncertain, the results maybe due to other extraneous variables. For example in an experiment trying to establish a relationship between diet and IQ, the extraneous variables might be parents IQ because intelligent parents might supply a better diet.
Ethical considerations of this method are:- Causal interference shouldn’t be made but often this misinterpretation of the data does happen. This is especially important with relation to socially sensitive issues such as IQ, which tend to rely on correlational data.
Interviews and questionnaire survey:- Come in many different forms. They vary in terms of the amount of structure and whether they are conducted face to face or require written answers. Non-direct interviews possess the least structure, with the person interviewed being free to discuss almost anything he or she wants. The role of the interviewer in non-directive interviews is to guide the discussion and encourage the interviewee to be more forthcoming. This type of interview is used very often in treatment of mental disorders. It provides rich in depth information, replication is not possible.
Informal interviews:- The interviewer listens patiently and focuses mainly on encouraging the interviewee to discuss issues in more depth or detail. There are certain general topics that the interviewer wishes to explore. This provides detailed information on specific topics, it is not possible to replicate.
Informal guided interviews:- It is informal but guided interviews it possess a little more structure than informal interviews. Before hand the interviewer identifies the issues to be addressed. During the course of the interview, further decisions are made about how and when to raise these issues. It provides detailed information on specific topics, it is difficult but not impossible to replicate.
Structural open-ended interviews:- Use a formal procedure in which all interviewees are asked precisely the same questions in the same order. Such procedure prevents the interviewee from side tracking the interview and taking control if it away from the interviewer. It is replicable. It permits good comparison between people.
Clinical interviews:- Used to assess patients with mental disorders. All of the interviewees are asked the same questions, but the choice of follow up questions depends on the answers that are given. This method was used by Paiget when talking to children.
Fully structured interviews:- A standard set of questions, it is asked in the same fixed order to all the interviewees and they are only allowed to choose their answers from a restricted set of possibilities. It is a face to face data gathering technique, could be conducted by phone or post. It is quick and easy to collect data.
Questionnaire surveys written questionnaires are completed by participants. It is replicable and it provides a lot of data quickly and cheaply.
In all interviews, there maybe problems of interviewer bias. It is also possible that people may not tell the truth in order to give a good impression which is known as social desirability.
Ethical considerations:- Confidentiality should always be respected, and especially where personal issues are involved. It may not be sufficient to simply withhold someone’s name since they may still be identifiable.
Interviewers should take great care to avoid material getting into the wrong hands. Interviewees should not feel obliged to answer questions that they find embarrassing and they should be reassured that they don’t have to answer any questions they don’t want to answer.
The research design:- When planning a research study there are various considerations to think about, to make the study successful when designing an experimental study the first step would be the research aims:- these are the stated intensions of question's that are planned to be answered. The aim is more general than the hypothesis. The aim tells us why a given study is being carried out, whereas the hypotheses tells us what the study is designed to test. For example an aim could be to see whether g term memory is influenced by the kind of processing that occurs at the time of learning.
A hypothesis is a formal unambiguous statement of what you believe to be true. The hypothesis is simply a prediction or expectation of what will happen in a given situation. We all form hypotheses all the time such as “my basket ball team will win today”. Psychologists state their hypothesis so that they will be clear about what they aim to prove or disprove in their research. Hypothesis is more specific for example “free recall from long term memory is higher when there is a semantic processing at the time of learning than when there is non- semantic processing.
There is also a hull hypothesis which is a statement of “no difference” or “no relationship” between the populations being studied. It consists of the predicted effect of the IV on the DV. For example a suitable null hypothesis could be :- “loud noise will have no effect on people’s ability to learn information”. There is the need of null hypothesis because it is much more precise, and because you can never prove something, it can only be disproved.
Variables:- Literally means anything whose value is free to change. Variables are usually divided up in to IV, DV and extraneous variables when used in an experimental design. The IV is the one which is specifically manipulated so that we can observe it’s effects on the DV. The DV is usually the one we are measuring or assessing. So for example in a study of the effects of stress on performance come participants are given an impossible puzzle to solve, where as others do an easy puzzle. In this case the IV is the difficulty of the puzzle and the DV is the performance in the task. Confounding variables are the variables that are controlled.
In non experimental research i.e. s, interviews and observations, a hypothesis is useful to have, but these will not identify a potentially causal relationship. For example an aim of an observational study might be “to study the feeding behaviour of cows”, observations are then made i.e. the process of designing research.
{Returning back to experimental research}:- There are two types of experimental hypotheses. A one tailed or directional hypothesis predicts the nature of the effect of the IV on the DV. For example “loud noise will reduce a person’s ability to learn information. A two tailed or a non directional hypothesis predicts that the IV will have an effect on the DV, but the direction of the effect is not specified. For example “loud noise will have an effect on a person’s ability to learn information.
there are three types of experimental design:- Independent design– each participant is selected for only one group, matched participants design– each participant is selected for only one group but the participants in the two groups are matched for some relevant factors such as age and sex and repeated measures design– each participant appears in both groups so that there are exactly the same participants in each group.
the advantages of independent groups design:-*There are no problems of order effects. *no participants are lost between the trials. *It can be used when a repeated measures design is inappropriate.
The disadvantages are:- *There may be important individual differences between participants to start with. *You need more participants than you do with a repeated measures design.
The advantages of matched participants design:- *it controls for some individual differences between participants. *Can be used when a repeated measures design inappropriate.
The disadvantages are:- *It is quite difficult to match pairs. *You need more participants than you do with a repeated measures design.
The advantages of repeated measures design:- *It controls for all individual differences. *It requires fewer participants.
The disadvantages are:- *It cannot be used in studies in which participation in one condition affects responses in the other. *Participants are likely to guess the purpose of the study, thus introducing problems with demand characteristics. *There are problems of order effects.
Order effects and counterbalancing:- when a repeated measures design is used there may be problems that result from participants doing the same task twice. The second time they carry out the task they may be:*better than first time because they have had a practise, *worse than first time because they are tires or bored.
The solution is called by a technique called counterbalancing:-*half the participants do condition A, *half the participants do condition B and followed by A.
Factors associated with research design:- In order for research to be conducted properly, there are certain guidelines that should be followed:-*standard instructions:-all participants are to be given identical instructions.
*standard procedures:-should be used for the collection of data, applies to all research. All participants are to be treated in exactly the same way.
*control of variables:- these are the variables that may mask the effect of the IV. Confounding variables must be controlled. Any variables that change between conditions such as participants becoming more tired or being more
motivated in one condition than another, are difficult to control. Such confounding variables are known asa constant error. It is important to control certain variables such as time of day and location, these are known as confounding variables. *operationalisation– some variables such as performance are very general so we need to decide in a precise way to measure them. For example performance could be measured in LTM by number of words remembered.
the advantages are:- generally provides a clear and objective definition of even complex variables.
the disadvantages:-use of operational definitions are entirely circular, often covers only part of the meaning of the variable, everyone is not always agreed on the accuracy of the operationalisation.
Pilot study:- this is a small scale study carried out before the main one in order to check procedures and design. It helps sort out problems allows for adjustments, thereby saving time and money.
Measuring and improving reliability:-one of the main goals of research is to design and carry out studies in such a way that replication of one’s findings is possible. In order to achieve that goal, it is important that measures we use should possess good reliability. Coolican 1994 points out “any we use in life should be reliable, otherwise it’s useless”.
Internal reliability:-refers to how consistent a method measures within it self. *split half technique:-used for establishing the internal reliability of the scores and compare them with the other half to see how similar they are , the same participant does both halves at the same time.
Inter-rater reliability:- used when determining the reliability of observations, to improve reliability in observations more than 1 observer must be used, all observers must be precise, clear categories of behaviour, all observers must be trained in the use of this system and inter-rater reliability can be measured using a correlational analysis.
External reliability:- this refers to how consistently a method measures over time when repeated. Results from an IQ test shouldn’t differ markedly from 1 occasion to another. This can be measured by using test –retest method, where the same test is given to participants on separate occasions to see if their scores remain relatively similar.
Measuring and improving the validity:-validity refers to whether a technique can achieve the purpose for which it was designed, valid means true. In a experiment one looks at whether the measurements taken will be a true reflection of what they are supposed to be measuring. 1 major problem with establishing validity in psychological research is that the more precisely and carefully you control conditions the less valid the measures are likely to be in terms of how they generalise to the real world