Revision List – Psychology Unit 1 (Research Methods) Assessment (Michaelmas 2) For this assessment, you could be asked to define any of the terms below, in relation to source material, and where relevant, to evaluate their use. Definition Strengths Weaknesses Independent variable An independent variable is the variable that is manipulated or changed by the researcher in an experiment to determine its effect on the dependent variable. /2 Dependent variable A dependent variable is the variable that is measured by the researcher to assess the effect of the independent variable. It represents the outcome or result of the experiment. /2 Operationalisation Operationalisation is the process of defining variables in a measurable way, so they can be objectively observed or quantified in a study. For example, measuring ‘stress’ by recording heart rate. /2 Extraneous variables Extraneous variables are variables other than the independent variable that could potentially influence the dependent variable and affect the outcome of an experiment. If not controlled, they can reduce the validity of the results by introducing unwanted variability. For example, noise levels or participants’ prior knowledge. /3 Experiment An experiment is a research method in which the researcher manipulates an independent variable to observe its effect on a dependent variable, while controlling extraneous variables. This method allows researchers to establish cause-and-effect relationships between variables. For example, testing how different levels of light affect participants’ reading speed. /4 Laboratory Experiment A laboratory experiment is an experiment conducted in a controlled, artificial environment, where the researcher manipulates the independent variable and measures the dependent variable, while strictly controlling extraneous variables. This ensures high internal validity. /2 High control over variables: Extraneous variables can be minimized, ensuring higher internal validity. Replicability: Procedures are standardised, making it easier to replicate the study and test for reliability. Low ecological validity: The artificial setting may not reflect real-life situations, reducing generalisability. Demand characteristics: Participants may guess the aim of the study and alter their behaviour, potentially biasing the results. Field Experiment A field experiment is an experiment conducted in a natural, real-world environment, where the researcher manipulates the independent variable and measures the dependent variable, while attempting to control extraneous variables as much as possible. /2 High ecological validity: Conducted in real-world settings, so finding is more generalisable to everyday life. Reduced demand characteristics: Participants may be unaware they are part of a study, leading to more natural behaviour. Less control over extraneous variables: This can reduce internal validity and make it harder to establish cause-and-effect relationships. Ethical issues: Participants may not give informed consent or be aware they are being studied, raising ethical concerns. Quasi Experiment A quasi experiment is a study where the independent variable is naturally occurring and not manipulated by the researcher. Participants are not randomly allocated to condition, as group membership is based on pre-existing characteristics e.g. gender or age. /2 High ecological validity: Since the independent variable is naturally occurring, the results are more likely to reflect real-world situations. Ethical advantages: It may be easier to study sensitive topics where random assignment is not possible or ethical e.g. studying the effects of gender or age. Lack of control over variables: Because the independent variable is not manipulated, it is harder to establish cause-and-effect relationships. Lower internal validity: Without random assignment, there may be confounding variables that influence the results, reducing the ability to draw clear conclusions. Independent groups design An independent groups design is an experimental design where different participants are assigned to each condition of the experiment. Each participant experiences only one condition, ensuring that there is no overlap between groups. /2 No order effects: Since participants are only exposed to one condition, issues like practice or fatigue effects are avoided. Simpler to organize: Each participant is assigned to only one condition, making the design easier to implement. Participant variables: Differences between participants in different groups e.g. age, background, may affect the results, reducing internal validity. More participants required: Since each condition requires a separate group of participants, this design can require more participants than a repeated measures design. Repeated measures design A repeated measures design is an experimental design where the same participants are exposed to all conditions of the experiment. Each participant serves as their own control, experiencing every condition, which allows for direct comparisons within individuals. /2 Fewer participants needed: Since the same participants are used in all conditions, fewer people are required compared to independent groups design. Controls for participant variables: As the same participants are involved in all conditions, individual differences e.g. age, gender, do not confound the results. Order effects: Participants may perform differently in later conditions due to factors like fatigue or practice which can affect the results. Demand characteristics: Participants may guess the purpose of the study after experiencing multiple conditions, potentially altering their behaviour. Matched Pairs Design A matched pairs design is an experimental design where participants are paired based on similarities in key characteristics e.g. age, IQ, and each member of the pair is assigned to a different condition of the experiment. /2 Controls for participant variables: By matching participants based on relevant characteristics, individual differences are minimized, improving internal validity. No order effects: Since participants are only exposed to one condition, there are no issues like fatigue or practice effects. Difficult and time-consuming: Finding suitable matches for participants can be complex and resources intensive. Matching may not be perfect: Even when participants are matched, it is difficult to account for all possible variables, meaning some differences may still influence the results. Structured Observation A structured observation is a type of observational research where the researcher uses a pre-determined system or checklist to record specific behaviours in a controlled and systematic way. /3 High reliability: The use of pre-determined system or checklist ensures consistency in how behaviours are recorded, making the study easier to replicate. Focused data collection: The researcher can concentrate on specific behaviours, improving the relevance and clarity of the data. Limited flexibility: Since behaviours are predefined, the researcher may miss important or unexpected behaviours that could provide valuable insights. Observer bias: The researcher’s expectations or the checklist may influence what is recorded, reducing objectivity. Unstructured Observation An unstructured observation is a type of observational research where the researcher records all relevant behaviours that occur during the observation without a pre-determined checklist or system. This allows for a more flexible and natural recording of data. /2 Rich and detailed data: The researcher can capture a wide range of behaviours and nuances that might be missed in more structured approaches. Flexibility: The researcher can adapt to the situation and observe unexpected or interesting behaviours as they occur. Low reliability: The lack of a standardised system for recording behaviours can make it harder to replicate the study or ensure consistency. Observer bias: The researcher’s personal interpretations or expectations may influence what they observe or record, affecting objectivity. Naturalistic Observation A naturalistic observation is a research method where the researcher observes behaviour in its natural environment, without manipulating or controlling any variables. /2 High ecological validity: Since behaviour is observed in a natural setting, the findings are more likely to reflect real-world behaviour. No participant bias: Participants are unaware they are being observed, reducing the likelihood of demand characteristics influencing their behaviour. Lack of control: The researcher has no control over extraneous variables, making it difficult to establish cause-and-effect relationships. Ethical issues: Observing participants without their knowledge can raise ethical concerns, such as issues with informed consent and privacy. Controlled Observation A controlled observation is a research method where the researcher observes behaviour in a structured environment where certain variables are manipulated or controlled. /2 High control over variables: Researchers can control extraneous variables, leading to more reliable and valid results. Easier replication: The structured nature of the observation allows for the study to be easily replicated by other researchers, improving reliability. Low ecological validity: The artificial setting may not reflect real-life situations, making it harder to generalise findings. Demand characteristics: Participants may behave differently because they are aware they are being observed in a controlled environment. Overt Observation Overt observation is a research method where participants are aware they are being observed. The researcher informs the participants of the study’s purpose and their role in the observation, ensuring transparency. /2 Ethical transparency: Since participants are aware they are being observed, it ensures informed consent and respects their rights. Less risk of ethical issues: There are fewer concerns about invasion of privacy compared to covert observation. Demand characteristics: Participants may alter their behaviour because they know they are being observed, which can affect the naturalness of their actions. Observer bias: The researcher may be influenced by knowing the participants are aware of the observation, potentially leading to biased interpretations or behaviours. Covert Observation Covert observation is a research method where participants are unaware, they are being observed. The researcher observes behaviour in a natural setting without the participants knowledge. /2 More natural behaviour: Since participants are unaware, they are being observed, their behaviour is likely to be more genuine and less influenced by the observer’s presence. No demand characteristics: As participants do not know they are part of the study, they are less likely to alter their behaviour in response to the observation. Ethical concerns: The lack of informed consent raises serious ethical issues related to privacy and deception. Observer bias: The researcher may interpret behaviours differently due to the lack of interaction with the participants or because they may be unable to ask clarifying questions. Participant Observation Participant observation is a research method where the researcher actively takes part in the group or activity being studied, while also observing and recording the behaviour of the participants. /2 In-depth insight: By actively participating, the researcher gains a deeper understanding of the group’s behaviour, attitudes, and interactions. Improved rapport: Being part of the group allows the researcher to build trust with participants, which can lead to more natural and authentic behaviour. Researcher bias: The researcher’s involvement in the group may influence their objectivity, leading to biased observations or interpretations. Ethical concerns: There may be issues with informed consent, deception, or lack of transparency, especially if the researcher is undercover or does not disclose their full role. Non-participant Observation Non-participant observation is a research method where the researcher observes the group or activity from outside without directly engaging or participating in it. /2 Objective observation: The researcher remains detached from the group, which helps reduce bias and allows for more objective data collection. No influence on behaviour: Since the researcher is not part of the group, they are less likely to alter the participants natural behaviour. Limited understanding: Without direct involvement, the researcher may miss key insights or fail to fully understand the context of participants’ behaviours. Difficulty in building rapport: The researcher may struggle to gain trust from participants, which could affect the authenticity of their behaviour. Time Sampling Time sampling is a technique used in observational research where the researcher records behaviours at predetermined time intervals. /2 Efficient data collection: By observing behaviour at set intervals, the researcher can collect large amounts of data in structured and manageable way. Reduces observer bias: The fixed intervals prevent the researcher from selectively observing behaviour, ensuring more systematic and objective data collection. May miss important behaviours: If a behaviour occurs outside the set time intervals, it may not be recorded, leading to incomplete data. Not fully representative: Behaviour that is highly variable might not be captured accurately, affecting the overall validity of the findings. Event Sampling Event sampling is a technique used in observational research where the researcher records every occurrence of a specific behaviour or event during the observation period. /2 Captures specific behaviours: This method ensures that behaviours of interest are recorded every time they occur, providing detailed data about those behaviours. Useful for infrequent behaviours: Event sampling is effective for observing rare or unusual behaviours that might be missed with other methods. May miss context: Focusing on specific events may lead to a lack of understanding of the broader context in which the behaviour occurs. Risk of observer bias: The researcher may place more focus on certain events or behaviours, which could lead to biased or incomplete data if other important behaviours are overlooked. Null Hypothesis A null hypothesis is a statement that suggests there is no significant effect or relationship between the variables being studies. It posits that any observed differences or effects are due to chance or random variation, rather than a true cause-and-effect relationship. /2 Alternative Hypothesis An alternative hypothesis is a statement that suggests there is a significant effect or relationship between the variables being studied. It proposes that any observed differences or effects are not due to chance but reflect a real cause-and-effect relationship. /2 One-tailed Hypothesis A one-tailed hypothesis is a type of hypothesis that predicts the direction of the effect or relationship between variables. It suggests that the independent variable will either increase or decrease the dependent variable, but not both. /2 Two-tailed Hypothesis A two-tailed hypothesis is a type of hypothesis that predicts a significant effect or relationship between variables, but without specifying the direction of the effect. It suggests that the independent variable could either increase or decrease the dependent variable. /2 Random Sampling Random sampling is a technique where every member of the target population has an equal chance of being selected for the study. This method aims to ensure that the sample is representative of the population, minimizing bias. Random sampling is typically achieved through methods like drawing names from a hat or using a computer-generated random selection process. /3 Reduces bias: Every participant has an equal chance of being selected, helping to avoid researcher bias and increasing the likelihood of a representative sample. Generalizability: A random sample is more likely to reflect the diversity of the target population, making the findings more applicable to a wider group. Difficult to implement: Obtaining a complete list of the population and selecting participants at random can be time-consuming and impractical in some situations. May still be unrepresentative: Even with random selection, the sample could still be biased due to chance, especially in small samples where certain groups may be overrepresented or underrepresented. Opportunity Sampling Opportunity sampling is a non-random sampling technique where the researcher selects participants who are easily accessible or available at the time of the study. This method often involves using people who are conveniently present, such as those who volunteer or are nearby. /2 Quick and easy: It is a fast and convenient method as participants are selected from those readily available, making it practical for researchers with limited time or resources. Low cost: Opportunity sampling typically requires less effort and fewer resources compared to other sampling methods, making it cost-effective. Bias and lack of generalizability: The sample may not be representative of the broader population, leading to biased results that cannot be easily generalized. Limited variety: The sample may consist of similar types of individuals (e.g., students in a specific area), reducing the diversity of the sample and affecting external validity. Self-selected Sampling Self-selected sampling is a sampling technique where participants volunteer to take part in a study, often in response to advertisements or invitations. This method relies on individuals choosing to participate based on their own interest or availability. /2 Easy to implement: Participants actively volunteer, which can make recruitment faster and simpler for the researcher. Ethical: Since participants choose to take part, informed consent is typically easier to obtain, ensuring ethical standards are met. Volunteer bias: Participants who volunteer may have characteristics or motivations that make them unrepresentative of the broader population, leading to biased results. Lack of generalizability: Because the sample may not be diverse or representative, findings may not apply to the wider population. Snowball Sampling Snowball sampling is a non-random sampling technique where initial participants are recruited through referrals from other participants. As each participant refers more people, the sample size grows like a snowball, making it useful for studying hard-to-reach or hidden populations. /2 Useful for hard-to-reach populations: It is particularly effective for studying groups that are difficult to access, such as specific subcultures or people with characteristics. Cost-effective: It often requires less time and resources compared to other sampling methods, as participants help recruit others. Bias: The sample may be unrepresentative because participants are more likely to refer people with similar characteristics, which reduces the diversity of the sample. Lack of generalizability: Due to the non-random selection, the findings may not be generalizable to the wider population. Definition Strengths Weaknesses Measure of central tendency A measure of central tendency is a statistic that represents the central or typical value in a set of data. The most common measures are the mean, median, and mode, which summarize the data by identifying a central point around which other values cluster. /2 Mean The mean is a measure of central tendency calculated by adding up all the values in a data set and then dividing the total by the number of values. It provides an average value, which represents the overall data set, but can be influenced by extreme values (outliers). /3 Uses all data points: Since it incorporates every value in the data set, the mean provides a comprehensive measure that reflects the entire set. Precise: It provides a specific, numerical value that can be used in further statistical analysis, making it useful for mathematical and inferential statistics. Affected by outliers: Extreme values (outliers) can significantly skew the mean, making it less representative of the data set in cases of skewed distributions. Not suitable for all data types: The mean is not appropriate for ordinal or nominal data, where other measures of central tendency (e.g., median or mode) might be more meaningful. Median The median is a measure of central tendency that represents the middle value in a data set when the values are arranged in numerical order. If there is an even number of values, the median is the average of the two middle values. /2 Not affected by outliers: Unlike the mean, the median is not influenced by extreme values, making it a more accurate measure of central tendency for skewed data. Works with ordinal data: The median can be used with ordinal, interval, or ratio data, making it versatile for different types of data. Less precise: The median does not take into account the actual values of the data, only their position, which can make it less precise than the mean, especially for symmetrical distributions. Not suitable for small data sets: In small data sets, the median may not provide a meaningful representation of the central tendency. Mode The mode is a measure of central tendency that represents the most frequent value or category in a data set. It is the value that appears most often, and it can be used with nominal, ordinal, interval, or ratio data. /2 Easy to calculate: The mode is simple to determine, especially in categorical data, and does not require complex mathematical operations. Works with any data type: The mode can be used with nominal data (e.g., categories or labels), where the mean and median may not be applicable. Unrepresentative with multiple modes: If there are multiple modes (bimodal or multimodal), it can be unclear which value truly represents the "central" tendency of the data. Ignores other data points: The mode only considers the most frequent value and does not account for the rest of the data, potentially overlooking important information. Measure of dispersion Measures of dispersion are statistics that describe the spread or variability of a data set. They show how much the data points deviate from the central value, indicating the degree of diversity within the data. Common measures of dispersion include the range, variance, and standard deviation, which help assess the consistency or variability of the data. /3 Range The range is a measure of dispersion that represents the difference between the highest and lowest values in a data set. It provides a simple indication of the spread or variability of the data. /2 Simple to calculate: The range is easy to compute, as it only requires identifying the highest and lowest values in the data set. Provides a quick overview: It offers a simple way to assess the spread of the data, especially when dealing with large datasets. Sensitive to outliers: The range can be heavily influenced by extreme values (outliers), which may give a misleading impression of the data's spread. Limited information: It only considers the two extreme values in the data set and ignores the distribution of the rest of the data, offering a less comprehensive measure of dispersion compared to others like standard deviation. Standard Deviation Standard deviation is a measure of dispersion that indicates the average amount by which data points differ from the mean of a data set. A lower standard deviation means the data points are close to the mean, while a higher standard deviation indicates greater variability. /2 Considers all data points: Unlike the range, standard deviation considers every value in the data set, providing a more comprehensive measure of variability. Accurate reflection of spread: It gives a more precise understanding of how data is spread around the mean, especially for normally distributed data. Affected by outliers: Extreme values (outliers) can significantly influence the standard deviation, making it less accurate for skewed data. Complex to calculate: Unlike simpler measures like the range or mode, standard deviation involves more complicated calculations, which may require statistical software or a calculator. Social Desirability Bias Social desirability bias is a type of response bias where participants alter their answers to appear more favourable or socially acceptable, rather than providing truthful or accurate responses. This can lead to distorted data, especially in surveys or interviews. /2 Pilot Study A pilot study is a small-scale preliminary study conducted before the main research to test the feasibility of the research design, methods, and procedures. It helps identify potential issues and allows for adjustments before conducting the full study. /2 Respect Respect, in the context of psychology research, refers to treating participants with dignity and consideration, ensuring their rights, privacy, and autonomy are upheld throughout the study. This includes obtaining informed consent and maintaining confidentiality. /2 Competence Competence, in the context of psychology research, refers to the researcher's ability to conduct the study effectively, ensuring they have the necessary skills, knowledge, and training to carry out the research responsibly and ethically. /2 Responsibility Responsibility, in the context of psychology research, refers to the researcher's duty to ensure the well-being and safety of participants, to conduct the research ethically, and to ensure the study complies with relevant guidelines and regulations. This includes minimizing harm and being accountable for the integrity of the research process. /2 Integrity Integrity, in psychology research, refers to the honesty and transparency of the researcher in conducting and reporting their study. This includes avoiding deception, ensuring the accuracy of data, and adhering to ethical guidelines throughout the research process. /2