Research & Methods Exam 1

Ontologies

Theories about the nature of reality (“what is”)

-Example: Is there a singular reality out there? Are there multiple realities continuously co-created by interactions with other entities?

Epistemologies

Theories of valid knowledge production: what constitutes legitimate knowledge? Is objective research possible?

-Example: Positivists, constructivists, and critical realists have different epistemologies

Methodology

Beliefs about how researchers can obtain valid knowledge about the world

-Example: Natural scientists would traditionally use remote sensing to conduct their studies

Positivism

A category within research paradigms. One reality independent of human thought where humans ultimately obtain truths about reality. This includes hypothesis testing and is used in disciplines including the physical sciences, biological and social sciences

-Example: Testing hypotheses about water quality contamination

Critical Realism

A category within research paradigms. One reality independent of human thought. The understanding that humans can only ever develop a partial understanding of reality, given that all knowledge is filtered through specific human contexts. The best method of investigation is a pragmatic choice based on the nature of the research questions. This is done mainly in social sciences and some natural sciences.

-Example: Mixed-methods research aimed at understanding key patterns of social-ecological relationships and their underlying causes.

Constructivism

A category within research paradigms. There are multiple realities where knowledge is created via dialectical relationships and subjective interpretation, often investigated through dialectical methods and qualitative data

-Example: Oral histories, open-ended interviews aimed at understanding national park visitors experiences

Operationalize

Move from a concept to a measurement/criteria

-Example: To operationalize a bird study within two different habitats, for example, you can set up trail cameras to track bird species and population/group size within these two different habitats

Quantitative methods

Collecting numerical data

-Examples: Collecting a student’s average hours of sleep each night and their end-of-semester GPA to see if more sleep correlates with a higher GPA

Qualitative methods

Collects data that will be analyzed as words

-Example: Interviewing fishermen on their fishing use— whether they usually fish from piers and elevated areas or on the beach

Mixed methods

A combination of quantitative and qualitative methods

-Example: Tracking people’s use of the recycling bins on campus (ex., # of recyclables recycled each week), interviewing people about their likelihood to recycle/their habits.

Research ethics

Ethical treatment of human and non-human research subjects

-Example: We earned our IRB certificates, which are supposed to uphold research ethics

Institutional review board

Institutional review board that ensures benefiance, justice, autonomy, and informed consent

-Example: when doing social surveys, it’s imperative that you follow these three guidelines to ensure your collected data is valid

Institutional Animal Care and Use Committee

(IACUC) Review board that approves conducting research with animals

-Example: If you were to test a hypothesis on the behavioral responses of lizards to different foods, you would need an IACUC approval form

Beneficence

Minimizing harm and maximizing benefits among participants

-Example: Giving participants chocolate after data collection

Justice

All subjects that volunteer/are selected are treated equally

-Example: All survey questions are the same for all subjects

Autonomy and informed consent

All subjects are informed of what they are answering and exactly what their data will be used for. They give their expressed consent to have their data used in the experiment

-Example: Before giving out a survey, make sure to reach out to participants, ensuring that they’re knowledgeable about the topic and 100% consent to using the data given by them

Empirical Approach

Data collected by observing or experiencing the real world

-Example: Walking around Fox Pond and writing down bird populations you observe

Data

Information collected for both qualitative and quantitative research

-Example: Tree diameter measurements; likert scale 1-5 agree-disagree response

Inductive approach

Bottom-up reasoning: find the data and let it speak to you

-Example: Finding students’ thoughts about the value of trees on campus

Deductive Approach

Top-down reasoning: going into research with a testable hypothesis

-Example: The water in Frenchman’s Creek is unsafe for swimming, water tests to prove it

Experimental Research

Studies testing for a reaction when a condition or important component is manipulated by the researcher (hypothesis possible but not required)

-Example: Mixing chemicals together in a chemistry lab (e.g., potassium and water)

Non-experimental Research

Studies testing for outcomes where no conditions are manipulated (hypothesis possible but not required)

-Example: Social surveys; convenience sampling of bird species around a pond

Hypothesis

Testable prediction about the results of your experiment

-Example: Bird population health will be greater in mangrove habitat vs. march grass

Theory

A more complex, unified explanation of observations that align with empirical evidence

-Example: Constructivism theory (people form understanding by interacting with the world and adjusting their thinking when new information conflicts with prior beliefs

Unit of analysis

Who or what is being analyzed? Who/what will you be able to draw conclusions about by the end of your study

-Example: In social surveys→ each individual person or group of peopleIn bio surveys → depends on the study, exs might be locations, particular observed species, etc.

Variables

The attributes of units that researchers measure. They VARY by unit

-Example: People vary in demographic characteristics and opinions Campus landscapes will vary in invasive flora and fauna

Categorical variables

Variables with mutually exclusive categories

-Example: Demographics, likert scale

Continuous variables

Variables that can be quantified with specific numerical values

-Example: measurements

Nominal variables

Categories with no inherent order, classified by name only

-Example: animal species

Ordinal variables

Contain categories with a rank order; but distances between categories are subjective

-Example: Education level

Interval Variables

Have a numerical value that can be measured along a scale

-IQ Test

Ratio variables

Same as interval, with added requirement that the number line has a zero value, which means none of that variable exists

-Example: pond depth

Measurement Level

The degree and nature of quantification implied by a particular type of measurement

-Example: (nominal, ordinal, interval, ratio —>)

Independent Variable

The predictor; the one you expect to influence or affect the other

-Example: type of soil which makes plants grow fastest

Dependent Variable

The outcome; the one you expect to be affected

-Example: the size of the plant

Directional Hypothesis

Predicted outcome of the relationship between variables

-Example: ES majors will be able to accurately identify more bird species than humanities majors

Non-directional hypothesis

No prediction of outcome but a non-directive awareness that there will be a difference between the variables

-Example: There will be a difference in native and non-native populations in dry and wet conditions

Null hypothesis

There is no relationship, effect, or difference between the variables you are studying. All statistical tests are testing against the null.

-Example: Grassy vs mangrove environments have no effect on bird health

Structured interview

Defined set of questions

-Example: Survey of bird knowledge on campus

Semi-structured Interview

Set of questions to start with room for spontaneous questions

-Example: Interview of fishers at local beaches with a couple questions to start conversation but no defined time limit and room for follow up questions based on interviewees

Unstructured Interview

No defined questions before interview, let direction be guided by interviewee

-Example: Going up to someone and starting a conversation and going with what they say to continue the discussion

Focus group

Researchers analyze discussion of group of people to draw conclusions about their interactions and reactions to questions

-Example: Dr. Boucquey overseeing student discussion on overfishing

Ethnography

Immersive research method where researcher experiences a culture/society to better understand their perspective

-Example: Living in an indigenous community

Participant observation

Researcher directly interacts with participants

-Example: Interviewing fisherman on a pier but would be fishing themselves

Non-participant observation

Observe from afar with no interactions with participants

-Example: sitting on a fishing pier and observing fishing dynamics

Case study

A specific example of a real-world event used to draw conclusions

-Example: Brazil paper

Grounded Theory coding

An inductive approach to research that goes from the ground up (ask question, categorize responses, develop themes), makes no hypothesis or assumptions about responses

-Example: “How well do you think people on campus manage to throw away their trash?”

Purposive coding

A deductive approach where researchers have a pre-established framework they are coding for

-Example: Does the interviewee mention “government intervention”? If so, provide a quote here:

Descriptive codes

Specific, local phrasing using direct quotes or words from data used in earliest stages of coding

-Example: “I HATE HOW THE GOVERNMENT TAKES ALL MY MONEY”

“Government taking” might be a descriptive code that could be used

Analytic codes

The “big picture” codes that describe conceptual themes within the data.

-Example: Concern for nature, geographic isolation, human-nature interactions

Coding Frame

A list of codes from the data being analyzed, iteratively organized

-Example: Recreation

      Recreational fishing

      Picnic table usage

      Tent camping

Maintenance

       Abundance of trash

       Quality of ammenities

Triangulation

A way of ensuring rigor in qualitative research, investigating if multiple sources, methods, and theories agree.

-Example: Bob the Builder agrees with Jamey Handorf and Peter Fritsche that Wireman Chapel is going to fall down (they all investigated independently, using different methods)

Participant Checking

Discussing with participants in a study to confirm their response after it is collected

-Example: People interviewed for a paper are sent the transcript of the discussion to double check before their interview is analyzed and added

Primary Research

Research conducted (data collected) and analyzed by the same researcher for one project

-Example: The Howell et al. paper conducted research about hydrology and invasive species and then analyzed it themselves.

Secondary Research

Original data that was collected by someone else. It can be qualitative or quantitative, and is usually cheap and fast. A disadvantage of secondary research is that it can be difficult to find recent studies, and many studies are not specific enough

-Example: The act of looking for previous data is secondary research.

Sampling

Selecting a smaller group from a larger population to represent the entirety of said population.

-Example: Selecting 10 students from ES 498-001 to represent the entire class.

Simple random sampling

Every item/person within a population has an equal chance of selection.

-Example: If we were taking a sample of students in ES 498-001, Hugh is just as likely to be selected as anyone in the class

Systematic sampling

Begin at random point, sample every nth case

-Example: The ants are in a line, close your eyes and randomly swing a stick, randomly stop the stick, that’s the ant you start with, then every other 3 or 5 or maybe 20 ants you sample

Stratified sampling

Divide population into categories or areas, then randomly sample within categories/areas

-Example:Divide the palm hammock into five zones, then randomly sample the trees in a randomly picked zone.

Purposive sampling

Sampling of specific people or places for a reason (not generalizable to a larger population)

-Example: You have a purpose for what you are studying, and hence the samplings you are using have been picked out purposefully.

Cluster sampling

A population is divided into clusters (groups), then everyone (or things) is sampled within a cluster

-Example: You have multiple bags of m’n’ms. Each bag is a cluster of m’n’ms that you study (randomly pick one bag and sample all m’n’ms in that bag)

Convenience sampling

Sampling of specific people or places because they are the most convenient (not generalizable to a larger population)

-Example: Whatever is the easiest and simplest way to sample.

Research measures

Physical or social instruments used to collect data

-Example: Survey, satellite tags (vertebrates), air quality sensors

Validity

The extent that accurately measures what is designed to be measured

-Example: How accurate is self reporting surveys

Reliability

The extent to which you can repeat the measure with the same results

-Video coding— repeatable by multiple researchers when using operational definitions (ethogram)

Judgemental validity

A way of assessing whether a measure is valid using experience

-Example: Dr.Frick is more equipped to research cetecans, while Dr.Highfill is more fit to research canids.

Predictive validity

Evaluating validity by making comparisons between a measure’s results and existing knowledge

-Example: How well does a new test compare to an already performed test? Ie. drone tracking animals when it was previously done on-ground

Internal Validity

How true are your conclusions about the relationships you are studying

-Example: sampling biases

External Validity

How does your case allow you to make generalizations about the broader world?

-Example: Is it possible for other labs to produce the same results if they follow it perfectly

Confounding variables

When the variable you are interested in as a cause of something could also be caused by something else

-Example: Time of day can cause more or less prey availability. Different types of weather can cause there to be less animal activity.

Statistical Validity

Are you using the right statistical tests for your data?

-Example: Categorical + categorical = chi square

Categorical + numerical = t-test or ANOVA

Numeric + numeric = correlation or regression

Numeric + categorical = Chi square (must put numeric IV  into categories)

Ordinal variables may be treated as categorical or numeric depending on the situation