Field Methods in Psychology

Descriptive Research Survey and Self 1. Planning Phase

Report 2. Data Collection Phase
3. Analysis Phase
is a method of research that seeks and 4. Interpretation Phase
describes something out there such as the 5. Reporting Phase
status, condition, or experience of a subject. It STRENGTHS AND LIMITATIONS
is the research design that is appropriate for
studies which aim to find out what prevails in Strengths
the present.
Descriptive research offers a detailed
Types and/or features of the design understanding of a topic, especially when
collected qualitatively. It can reveal unexpected
Descriptive research has several types. These insights, generate new research questions, and
are the surveys, case studies, trend studies,
help from hypotheses.
content analysis, feasibility studies,
development studies, follow-up studies, Limitations
evaluation studies, ethnographic studies,
relational studies, ex post facto studies, and Descriptive research can't establish cause and
effect. For instance, if a student believes a
replication and secondary analysis.
learning strategy helped them, we can't
Key features are: conclude it actually did. We also must be
cautious of reactivity, where people change
• Standardized Questionnaires: their behavior when observed, or provide
• Close and Open-ended Questionnaires socially desirable answers in surveys, like
• Large Sample Size parents denying smoking during pregnancy to
• Anonymity and Confidentiality conform to social norms.

Steps/Procedure of the Design

Observational Research Design Group
1. Identify the Research Problem
2. Define the Population and Sample
3. Design the Survey or Self-Report Tool A key method within this design, involves
4. Pre-test the Survey or Self-Report Tool observing and recording events or behaviors
5. Distribute the Survey or Self-Report without interfering or altering the environment.
Tool It provides detailed and accurate data, helping
6. Collect Responses researchers understand “what is” rather than
7. Analyze the Data - Quantitative Analysis “why it happens”.
or Qualitative Analysis
Types of Observational Research
8. Interpret the Results
9. Report the Findings 1. Participant observation – the
10. Follow-Up Actions (Optional) researcher actively participates in the
environment being studied
Procedure:
Field Methods in Psychology
2. Nonparticipant observation - the Limitations
researcher observes participants
without engaging or influencing the • Potential for bias
environment • Difficulty establishing causality
3. Structured observation – uses a • Generalizability challenges
predefined framework or checklist to • Ethical considerations
systematically record specific behaviors • Time- consuming
4. Unstructured observation – does not
Correlational Research Study
follow a strict framework; the
researcher observes behaviors freely A correlational study is a research method used
5. Covert observation – the participants to examine the relationship between two or
are unaware they are being observed more variables without manipulating them. This
6. Overt observation – are aware that type of study aims to identify whether a
they are being observed relationship exists, its direction (positive,
7. Direct observation – involves watching negative, or no correlation), and its strength.
subjects as they perform tasks or However, it does not establish causation, as it
engage in activities in time only observes and measures naturally occurring
8. Indirect observation – this method associations.
involves studying the results or traces
of behavior rather than behavior itself Types/Features of the Design

Steps/Procedure of the Design 1. Positive correlation - variables move in

the same direction (e.g., as one
1. Set a goal increases, the other increases).
2. Choose the observation type 2. Negative Correlation - variables move
3. Choose your sample in opposite directions (e.g., as one
4. Prepare data collection tools increases, the other decreases).
5. Conduct the observation 3. Zero Correlation - no identifiable
6. Record observation and data relationship between the variables.
7. Analyze the data
8. Interpret findings Key Features:

• Involves quantitative data.

• Relies on statistical tools like Pearson’s
STRENGTHS AND LIMITATIONS correlation coefficient or Spearman’s
rank correlation.
Strengths
• Non-causal: it examines relationships
• High ecological validity but does not imply causation
• Exploration of complex phenomena
• Hypothesis generation
• Cost-effective and efficient
• Flexibility and adaptability
Field Methods in Psychology
Steps /Procedures of the Design: 2. Third variables or Sample bias: Hidden
variables may influence relationships
1. Identify Research Variables: Select the and affect generalizability.
independent and dependent variables to
study. 3. Limited Control Over Variables: harder
to determine the true nature of the
2. Formulate a Research Hypothesis: Create a relationship between the variables
statement predicting the relationship between under study
the variables.

3. Collect Data: Use surveys, existing datasets, Comparative Research Design

observations, or experiments. Comparative analysis means describing and
explaining the similarities and differences of
4. Analyze Data: Use statistical methods to situations or consequences among large-scale
calculate the correlation coefficient (e.g.,
of social units such as regions, nations,
Pearson r). societies, and cultures.
5. Interpret Results: Assess the strength and This definition reflects traditions such as cross-
direction of the relationship and consider cultural analysis in anthropology, cross-societal
limitations.
analysis in sociology, cross-national analysis in
Strengths and Limitations political science, comparative historical analysis
in history, and psychological analysis.
Strengths
Key Features of Comparative Research Design
1. Efficient for identifying relationships
between variables in natural settings. Focus on comparison - act of comparing two or
more things with a view to discovering
2. Useful in early stages of research to something about one or all of the things being
generate hypotheses. compared

3. Can analyze large data sets. No causality or manipulation - no intention of

establishing/proving a “cause and effect”
4. Non-invasive, no need for experimental
relationship and is done in a non-manipulative
manipulation.
setting.
5. Ethical and practical considerations,
Selection of cases, groups & variables -
provide valuable insights without
selection process depends on research
requiring experimental control.
objectives.
Limitations
Types of comparison - can be cross-national or
1. Inability to establish causation: do not crossgroups comparisons.
control for extraneous variables that
may influence the results, cannot
establish cause-and-effect.
Field Methods in Psychology
Types of Comparative Research Design Fostering Critical Thinking and Problem-Solving
Skills: Researchers are compelled to question
1. Cross-sectional - comparing different preconceived notions, analyze diverse
groups at a single point of time perspectives, and develop nuanced solutions
2. Longitudinal - comparing same groups based on empirical evidence. This approach
overtime to observe changes fosters a more open-minded and rigorous
3. Case-oriented - detailed comparison of approach to inquiry, paving the way for
small number of cases often to explore creativity and innovation
complex phenomena
4. Variable-oriented - comparing large Limitations
number of cases with a focus on
identifying patterns or relationships Cultural and Methodological Biases:
between variables Researchers' own cultural biases can influence
the selection of cases, the interpretation of
Steps/Procedures of the Design data, and the framing of research questions.
Methodological biases can also arise from
1. Define objectives and score differences in research practices, sample sizes,
2. Gather relevant data and information or data collection techniques across different
3. Select appropriate criteria for cultural settings.
comparison
4. Establish a clear framework Oversimplification of Complex Phenomena:
Comparative research, by its nature, seeks to
identify similarities and differences. This focus
STRENGTHS AND LIMITATIONS can sometimes lead to an oversimplification of
complex social phenomena, ignoring the
Strengths nuanced and multifaceted nature of the
subjects under study. This simplification can
Enhanced Understanding and Broader
result in a loss of depth and richness in
Perspectives. By examining multiple cases or
understanding the phenomena.
subjects in parallel, researchers gain a broader
perspective and deeper insights into their study Language: Language is one of the most
area. This approach helps uncover underlying important barriers in comparative studies, i. e.
patterns and similarities across different cases, the meaning of particular terms may be lost in
which might remain unnoticed in single-case the translation of data, and also the same term
studies. may have different meanings in different
contexts. The language barrier can also affect
Cross-Cultural Insights: Comparative research
the validity of data collection tools like
delves into the diverse traditions, norms, and
questionnaires
values that distinguish one society from
another, illuminating the intricate mosaic of
human cultures. These types of studies often
bring together researchers from different
backgrounds and disciplines.
Field Methods in Psychology
Ex Post Facto Design • The research tries to analyze the ‘how’
and ‘what’ aspect of an event
- is a kind of research in which the
• Explores possible effects and causes
researcher aims to investigate possible
causes behind an effect that has already Types of Ex Post Facto
occurred.
- ex post facto came from the Latin word Ex Post Facto research designs can be
means “after the fact” categorized into two main types:
- it is sometimes called causal
Correlational studies - studies that examine the
comparative because its purpose is to
relationship between two or more variables to
investigate cause and effect
identify patterns or associations without
relationship between independent and
implying causation.
dependent variables
- researcher would try to find out the Criterion group studies - studies that compare
various events that have occurred and groups that differ based on a specific criterion
the many possibilities that could have to assess differences in outcomes. Example:
contributed to the concerned effect individuals who have experienced a particular
- an ex-post facto research can be event versus those who have not.
defined as an empirically based
investigation which does not involve Non-Manipulative Approach - The researcher
the researchers’ direct control over the does not manipulate the independent variable;
independent variables because they instead, they observe and analyze existing
have already led to effects which can no conditions or events that have already taken
more be manipulated. place. Thus, the name "ex post facto."
- The term ex-post facto according to Causal Relationships - These studies generally
Landman (1988: 62) is used to refer to do not establish definitive cause-and effect due
an experiment in which a researcher, to the lack of random assignment and
instead of finding a treatment, manipulation.
examines the effect of a naturally
occurring treatment after it has Grouping Based on Pre-existing Characteristics
occurred - Participants are grouped based on
characteristics that exist prior to the study, such
Characteristic of Ex Post Facto as age, gender, or health status.
• The research has a control or a Control Groups - Researchers often use control
comparison group groups to compare outcomes with experimental
• The behavior, action, event or the groups. Which helps with analyzing the effects
treatment or the independent variable of the independent variable on the dependent
of the research cannot be manipulated variable, even though no manipulation occurs.
or changed
• The research focuses on the effects Data Collection Techniques - Common methods
for data collection in ex post facto research
Field Methods in Psychology
include surveys, interviews, and analysis of Limitations
existing records or literature.
Can't Prove Cause-and-Effect: You can't be sure
Steps/Procedures of the Design that one thing causes another because you
can't control or manipulate variables.
1. Identify the Research Problem: Define
the specific event or condition to study. Confounding Variables: Other factors might
2. Select Participants: Choose subjects influence the results that you can’t account for,
based on existing characteristics making it hard to draw clear conclusions
relevant to the study
3. Collect Data: Gather information on the Bias in Data: If you're using past data, people's
independent and dependent variables. memories or records might not be accurate,
4. Analyze Data: Use statistical methods which can skew results.
to assess relationships between No Control Over Timing: You can't control
variables. when or how things happen, so it might be
5. Interpret Results: Draw conclusions harder to link cause and effect.
about the causal relationships identified
Limited Applicability: Because the data isn't
randomly chosen, the results might not apply to
STRENGTHS AND LIMITATIONS everyone.

Strengths Longitudinal Studies

Ethical and Practical: It's useful when Longitudinal studies are primarily a qualitative
experimenting isn't possible or ethical (e.g., research method because the researcher
studying health effects of smoking). observes and records changes in variables over
an extended period.
Real-Life Data: It studies real-world situations,
making the findings more applicable to Types/Features of the Design
everyday life. Panel study - the researcher uses data
Preliminary Insights: It’s helpful when you can't collection methods like surveys to gather
conduct an experiment but still want to information from a fixed number of variables at
understand relationships between variables. regular but distant intervals, often spinning into
a few years.
Cost-Effective: Since it uses existing data, it
saves money on data collection. Retrospective study - the researcher depends
on existing information from previous
Studying Non-Manipulable Variables: Useful systematic investigations to discover patterns
for studying things you can't control or leading to the study outcomes. In other words,
manipulate, like age or gender. a retrospective study looks backward.

Cohort study - a cohort study entails collecting

information from a group of people who share
Field Methods in Psychology
specific traits or have experienced a particular Steps/Procedures of the Design
occurrence simultaneously.
1. Define research question and objectives
Features: 2. Choose study design (cohort, panel,
trend)
1. Repeated Measurements: Data is collected 3. Select participants/data sources
at multiple time points to track changes or 4. Develop data collection tools
trends over time. 5. Pilot test
2. Long Duration: These studies can span years 6. Collect baseline data
or even decades, providing insight into long- 7. Conduct follow-up measurements
term effects or outcomes. 8. Analyze data
9. Interpret results
3. Cohort Groups: Participants are often 10. Report findings.
grouped by shared characteristics (e.g., age,
disease status) and tracked over time to STRENGTHS AND LIMITATIONS
observe how these characteristics influence Strengths
future outcomes.
1. Shows cause-and-effect relationships
4. Cause-and-Effect Relationships: Because of
the temporal aspect, longitudinal studies help 2.Examines changes over time
researchers examine potential cause-and-effect
relationships between variables. 3. Provides detailed insights

5. Data Collection: Typically involves collecting 4. Helps predict future outcomes

various types of data, such as medical records, 5. Validates theories
surveys, interviews, or biological samples.
6. Informs policy decisions
6. Retention Challenges: Keeping participants
involved over long periods can be difficult, and Limitations
attrition (dropout) is a common issue.
1. Time-consuming
7. Flexibility: Researchers can adjust the study’s
2. Expensive
focus over time, allowing for the exploration of
new hypotheses as data are gathered. 3. Participants may drop out
8. Observational Nature: Longitudinal studies 4. Data quality issues
are usually observational rather than
experimental, meaning they don’t manipulate 5. Biased sample selection
variables but observe natural occurrences.
6. Difficulty analyzing complex data

7. Ethical concerns
Field Methods in Psychology
Cross-Sectional Studies Cross-sectional studies allow you to collect data
from a large pool of subjects and compare
Cross-sectional study is a type of observational differences between groups
study that analyzes data from a population, or a
representative subset, at a specific point in Cross-sectional studies capture a specific
time. Think of it as taking a "snapshot" of a moment in time. National censuses, for
group of people at one particular moment instance, provide a snapshot of conditions in
that country at that time.
Types/Features of the Design
Limitations
Descriptive cross-sectional studies - aims to
describe the characteristics of a population at It is difficult to establish cause-and-effect
one point in time relationships using cross-sectional studies, since
they only represent a onetime measurement of
Analytical cross-sectional studies - aims to
both the alleged cause and effect.
investigate the associations between variables
within a population at one point in time. Since cross-sectional studies only study a single
moment in time, they cannot be used to
Steps/Procedures of the Design analyze behavior over a period of time or
1. Define Research Question and establish long-term trends.
Objectives: Clearly state what you want The timing of the cross-sectional snapshot may
to investigate, including the specific be unrepresentative of behavior of the group as
variables and their relationships. a whole
2. Choose Population and Sample: Select
a representative subset of the target Cohort Study
population using appropriate sampling
methods. A cohort study is a type of observational study
3. Design Data Collection Instrument: that follows a group of participants over a
period of time, examining how certain factors
Develop valid, reliable, and ethical tools
(e.g., surveys, interviews) for gathering (like exposure to a given risk factor) affect their
data. health outcomes.
4. Collect and Analyze Data: Types/Features of the Design
Systematically gather data at one point
in time and analyze using descriptive Prospective cohort studies
and inferential statistics
involves recruiting a group of participants and
STRENGTHS AND LIMITATIONS following them over time to gather new data

Strengths look forwards in time, to examine the

relationship between the exposure and the
Because you only collect data at a single point outcome.
in time, cross sectional studies are relatively
cheap and less time consuming than other
types of research.
Field Methods in Psychology
Retrospective cohort studies Due to their long nature, cohort studies are a
particularly good choice for studying rare
involve using preexisting data exposures, such as exposure to a new drug or
look backwards in time to examine the an environmental toxin. Other research designs
relationship between the exposure and the aren’t able to incorporate the breadth and
outcome. depth of the impact as broadly as cohort studies
do.
Steps/Procedures of the Design
Because cohort studies usually rely on large
The procedure for a cohort study involves the groups of participants, they are better able to
following steps: control for potentially confounding variables,
such as age, gender identity, or socioeconomic
1. Select a cohort: Identify a group of people
status. Relatedly, the ability to use a sampling
who meet the study's selection criteria and
method that ensures a more representative
are free of the outcome being studied. The
sample of the population leads to findings that
participants should have common
are typically much more generalizable, with
characteristics, except for their exposure
higher internal validity and external validity.
status.
Limitations
2. Measure exposure: Obtain baseline data on
the exposure to risk factors. Cohort studies can be extremely time-
consuming and expensive to conduct due to
3. Divide into groups: Divide the participants
their long and intense nature.
into two groups: one group is exposed to
the risk factor, and the other is not. Cohort studies are at risk for biases inherent to
long-term studies like attrition bias and
4. Follow up: Follow the participants over
survivorship bias, as participants are likely to
time and measure the outcomes.
drop out over time. Measurement errors like
5. Analyze data: Analyze the data to estimate omitted variable bias and information bias can
the risk of disease associated with exposure also confound your analysis, leading you to
draw conclusions that may not be true.
STRENGTHS AND LIMITATIONS
Like many other experimental designs, cohort
Strengths studies can raise questions regarding ethical
considerations. This is particularly the case if
Cohort studies are better able to approach an
the exposure of interest is harmful, or if there is
estimation of causality than other types of
no known treatment for it. Prior to beginning
observational studies. Due to their ability to
your research, it is critical to ensure that
establish temporality, multiple outcomes, and
participation in your study is fully voluntary,
disease incidence over time, researchers are
informed, and as safe as it can be for your
able to determine with more certainty that the
research subjects.
exposure indeed preceded the outcome. This
strengthens a claim for a cause-and-effect
relationship between the variables of interest.
Field Methods in Psychology
Meta Analysis The confidence interval for the summary effect
only considers within-study error (sampling
Meta-analysis is a statistical procedure used to error)
combine and synthesize findings from multiple
independent studies to estimate the average Random-Effect Model
effect size for a particular research question.
The random-effects model assumes that the
Meta-analysis goes beyond traditional narrative true effect size can vary between studies. This
reviews by using statistical methods to variation could be due to differences in study
integrate the results of several studies, leading populations, interventions, or other
to a more objective appraisal of the evidence. unmeasured factors. The studies included in the
meta-analysis are considered a random sample
This method addresses limitations like small from a distribution of possible effect sizes. The
sample sizes in individual studies, providing a random-effects model is generally more
more precise estimate of a treatment effect or appropriate when:
relationship strength.
• Studies are collected from the
Types/Features of the Design
published literature, where variations in
Fixed-Effect Model study characteristics are expected.
• The researcher wants to generalize the
The fixed-effect model assumes that all studies findings to a broader range of
in a meta-analysis share a single, common true populations and settings
effect size. Any observed differences in effect
sizes between studies are attributed solely to Features:
sampling error. This model is best suited for
Study weights are calculated by considering
situations where:
both within-study variance and between-
• The researcher believes that all studies variance
included studies are functionally
The confidence interval for the summary effect
identical, meaning all factors that could
is wider because it accounts for both within-
influence the effect size are the same in
study error and between-studies variance.
all studies.
• The goal is to estimate the common Steps/Procedures of the Design
effect size for a specific, narrowly
defined population, without Step 1: Define a Research Question
generalizing to other populations. Step 2: Develop a Search Strategy
Features: Step 3: Search the Literature
Study weights are assigned based on the Step 4: Screen and Select Research Articles
inverse of the study's variance, with larger
studies receiving more weight. Step 5: Evaluate the Quality of Studies

Step 6: Choose the Effect Size

Field Methods in Psychology
Step 7: Assess Heterogeneity results. This can lead to an overestimation of
the true effect size. Sources suggest addressing
Step 8: Choose the Meta-Analytic Model this bias by searching for unpublished studies,
Step 9: Perform the Meta-Analysis using statistical techniques to detect and adjust
for publication bias, and conducting sensitivity
Step 10: Conduct Sensitivity Analysis analyses.

Strengths and Limitations Heterogeneity across studies: Variation in study

design, population characteristics,
Increased statistical power: Meta-analysis
interventions, and outcome measures can lead
combines data from multiple studies, which
to heterogeneity in effect sizes. While random-
effectively increases the overall sample size.
effects models can account for this
This increased statistical power enhances the
heterogeneity, excessive variability can limit the
ability to detect statistically significant effects
interpretability and generalizability of the
that may not be apparent in individual studies,
findings. Careful consideration of the sources
especially when those studies have small
and potential impact of heterogeneity is crucial.
sample sizes.
Dependence on the quality of included studies:
Improved precision of effect size estimates: By
The validity of a meta-analysis is contingent on
pooling data from multiple studies, meta-
the quality of the individual studies included. If
analysis provides more precise estimates of the
the included studies are poorly designed or
effect size of a particular treatment or
executed, the findings of the meta-analysis will
intervention. This is particularly valuable when
be unreliable. It's essential to critically appraise
individual study results are inconsistent or
the methodological quality of studies and use
contradictory, allowing for a more robust and
appropriate quality assessment tools.
reliable estimation of the true effect size.
Time and resource intensive: Conducting a
Enhanced generalizability of findings: Meta-
meta-analysis is a complex and time-consuming
analysis allows for the examination of the
process, requiring extensive literature searches,
consistency of findings across a range of studies
data extraction, and statistical analysis.
conducted in different settings, with diverse
populations, and using various methodologies. Non-experimental Predictive Study
This provides a more comprehensive
understanding of the phenomenon being A research methodology that aims to predict
investigated and enhances the generalizability outcome/relationship w/out manipulating
of the findings to a broader population. variable.

Limitations Focuses on examining existing data and

variables to make prediction/identify
Publication bias: Meta-analysis can be association between them.
susceptible to publication bias, which occurs
when studies with statistically significant or Types/features of the Design
positive results are more likely to be published
1. Correlational design
than studies with non-significant or negative
Field Methods in Psychology
- Examines the strength and direction of Steps/Procedure of the Design
relationships between two or more
variable. Pre-Research Phase
- Utilizes correlation coefficient ▪ Define research objectives: Identify
(Pearson’s r). predictive goals and research questions.
- Cannot determine causation, only
association. ▪ Conduct literature review: Examine
2. Regression Analysis existing research on the topic.
- (Multiple regression) can handle
▪ Develop conceptual framework:
multiple independent variable
Visualize relationships between
- Extends correlation by modeling
variables
relationship between dependent and
independent variable. ▪ Formulate hypotheses: Predictive
- Provides equation that can be used for statements based on literature review
predictions.
3. Longitudinal Design (Predictive ▪ Select study design: Choose predictive
longitudinal study) non-experimental design (prospective,
- Collects data from the same subject retrospective, cross-sectional).
repeatedly over a long period.
Research design phase
- Use to study how variable change over
time and make prediction about future ▪ Identify variables: Select predictor
outcomes. (independent) and outcome
4. Cross-sectional design with Predictive (dependent) variables
elements
- Collects data at a single point in time ▪ Determine data collection methods:
from different subjects. Surveys, observations, existing records,
- can be used for predictive purposes by or secondary data
looking at association between
▪ Select sampling strategy: Non-
variables within that snapshot.
probability (e.g., convenience) or
- cannot track changes overtime but can
probability sampling (random)
identify relationship that are useful for
prediction ▪ Determine sample size: Calculate
5. Ex Post Facto Design (Retrospective required sample size for statistical
Design) power
- Studies existing data or phenomena
without any manipulation of variables. ▪ Develop data collection tools:
- Often involves looking back at historical Questionnaires, observation protocols,
data to find predictors of an outcome. or data extraction forms.
- Useful when experiments cannot be
conducted due to ethical or practical
limitations.
Field Methods in Psychology
Strength and Limitations

Strengths

Practicality: No controlled environments.

Real-world relevance: Studies real-world

scenarios

Time-efficient: Faster data collection

Limitations

Causality-Cannot establish cause-and-effect.

Confounding variables- Uncontrolled variables

may influence results.

Generalizability-Results may not be universally

applicable.