Non-scientific methods
Selectively observing -> Remembering and seeing things that agree with our beliefs when casually
observing.
Scientific method
1. Empirically testable (observations that are tested)
2. Replicable
3. Objective (shouldn’t matter who is performing the study, everyone should get the same
result, so no room for subjective interpretation)
4. Transparent (publicly sharing of relevant information)
5. Falsifiable (there must be data imaginable that it can be disproven)
6. Logically consistent/coherent (no internal contradictions)
Scientific claims
Observation -> doesn’t describe/explain general relation (can be accurate or inaccurate)
Hypothesis -> a statement that describes/explains a pattern/general relation (can be not or strongly
supported)
Law -> Type of hypothesis, which is really precise (for example mathematical) and strongly supported
Theory -> Overarching explanation of many related phenomena. Most well established explanation,
because they consist of hypothesis that have survived the scrutiny of the scientific method (trying to
refute hypothesis)
Goals
Universalistic research -> tries to provide explanations that apply generally (to all groups, all people
or societies, and every ‘form’)
Particularistic research -> aimed at describing or explaining a phenomenon that occurs in a specific
setting, or concerns a specific group (specific location, specific time)
Applied research -> Directly aimed at solving a problem (often particularistic and can also provide
fundamental knowledge)
Fundamental research -> To further our understanding of the world around us, without an
immediate application. (almost always universalistic)
Quantitative/Qualitative research
Empirical Cycle
Empirical cycle -> The process of coming up with hypotheses and testing these against empirical data
in a systematic and rigorous way
Observation phase (Sparks idea for hypothesis) -> Induction [phase (the observed relation is turned
into a general rule) -> Deduction phase (hypothesis is transformed to prediction) -> Testing phase
(summarizing data to statistics) -> Evaluation phase (interpret results)
Like this hypotheses can be specified
,No scientific statement can ever be proven once and for all. The best we can do is produce
overwhelming support for a hypothesis. There are often alternatives for a failure to confirm, rather
than just a rejection of the hypothesis (think of a methodological issues).
Criteria
Reliability -> If you repeat the study and then find consistent results
Validity -> If the conclusion accurately reflects reality
- Internal validity -> Relevant when the hypothesis describes a causal relationship. If the
observed effect is actually due to the hypothesized cause
- External validity -> If the hypothesized relationship also holds in other settings and other
groups (generalization)
- Construct validity -> Whether our methods actually reflect the properties we intended to
manipulate and measure
o Face validity -> That quality of an indicator that makes it seem a reasonable measure
of some variable.
o Criterion-related (predictive) validity -> The degree to which a measure relates to
some external criterion. The validity of a written driver’s test is determined, in this
sense, by the relationship between the scores people get on the test and their
subsequent driving records.
o Convergent/Discriminant validity -> Seeing whether the scores relate to similar and
different variables in a way that we expect.
An approach van be the multi-trait multi-method matrix approach: Use
different instruments to measure two traits. (Like cat fondness and pizza
fondness) -> conclude: no association between different constructs
measured with different methods.
You can see which measure instrument has low construct validity when the
measurements are not as expected.
o Content validity -> The degree to which a measure covers the range of meaning
included within a concept. (a math test needs to include the different parts)
Causal relationship
Hume’s criteria
1. Cause and effect are connected
2. Cause precedes the effect
3. Cause and effect occur together consistently
4. Alternative explanations can be ruled out
Introduction
Epistemology -> The science of knowing; systems of knowledge
Methodology -> The science of finding out; procedures for scientific investigation
Agreement reality -> Those things we “know” as part and parcel of the culture we share with those
around us
Errors:
, 1. Inaccurate observation
2. Overgeneralization
3. Selective observation
4. Illogical reasoning
a. Gambler’s fallacy (assuming a consistent run of either good or bad luck foreshadows
its opposite)
Three objections in regard to social regularities:
1. Some may seem trivial (“everyone knows that”)
2. Contradictory cases may be cited (indicating that the “regularity” isn’t totally regular)
a. An exception doesn’t always make a statement less true (woman earn less)
3. The people involved in the regularity could upset the whole thing if they wanted to
a. Does not happen enough for a serious threat
Variables of interest:
- Variables can cause other variables
- Independent variable: A variable with values that are not problematic in an analysis but are
taken as simply given. Presumed to cause or determine a dependent variable
- Dependent variable: A variable assumed to depend on or be caused by another.
Attributes -> Characteristics or qualities that describe an object
Construct -> Property in general, abstract
Variable -> Specific, concrete expression of the construct as measurable or manipulable.
Categorical variable -> Distinguish either nominal or ordinal categories . Does not make sense to
calculate a mean. Rather use the frequency or statistics.
- Binary/dichotomous variable: distinguishes between two categories.
- Polytomons: more categories
Quantitative variable -> Allow us to determine not only that people differ on a certain property, but
also to what extent they differ (for interval and ratio). It makes sense to calculate a mean
Discrete variable -> Can only take on a limited set of values. (Nominal and ordinal, yet quantitative
can also be discrete)
Continuous variable -> Always possible, in theory, to find a value between any other two values.
(between 6,5 and 6,4, there is 6,45)
3 Major purposes of social research:
- Exploration
- Description
- Explanation
Variables of disinterest
- Confounder/Lurking variable: Is related to both the independent and dependent variable
and partially, or even entirely, accounts for the relationship between these two. Is not
measured
Selectively observing -> Remembering and seeing things that agree with our beliefs when casually
observing.
Scientific method
1. Empirically testable (observations that are tested)
2. Replicable
3. Objective (shouldn’t matter who is performing the study, everyone should get the same
result, so no room for subjective interpretation)
4. Transparent (publicly sharing of relevant information)
5. Falsifiable (there must be data imaginable that it can be disproven)
6. Logically consistent/coherent (no internal contradictions)
Scientific claims
Observation -> doesn’t describe/explain general relation (can be accurate or inaccurate)
Hypothesis -> a statement that describes/explains a pattern/general relation (can be not or strongly
supported)
Law -> Type of hypothesis, which is really precise (for example mathematical) and strongly supported
Theory -> Overarching explanation of many related phenomena. Most well established explanation,
because they consist of hypothesis that have survived the scrutiny of the scientific method (trying to
refute hypothesis)
Goals
Universalistic research -> tries to provide explanations that apply generally (to all groups, all people
or societies, and every ‘form’)
Particularistic research -> aimed at describing or explaining a phenomenon that occurs in a specific
setting, or concerns a specific group (specific location, specific time)
Applied research -> Directly aimed at solving a problem (often particularistic and can also provide
fundamental knowledge)
Fundamental research -> To further our understanding of the world around us, without an
immediate application. (almost always universalistic)
Quantitative/Qualitative research
Empirical Cycle
Empirical cycle -> The process of coming up with hypotheses and testing these against empirical data
in a systematic and rigorous way
Observation phase (Sparks idea for hypothesis) -> Induction [phase (the observed relation is turned
into a general rule) -> Deduction phase (hypothesis is transformed to prediction) -> Testing phase
(summarizing data to statistics) -> Evaluation phase (interpret results)
Like this hypotheses can be specified
,No scientific statement can ever be proven once and for all. The best we can do is produce
overwhelming support for a hypothesis. There are often alternatives for a failure to confirm, rather
than just a rejection of the hypothesis (think of a methodological issues).
Criteria
Reliability -> If you repeat the study and then find consistent results
Validity -> If the conclusion accurately reflects reality
- Internal validity -> Relevant when the hypothesis describes a causal relationship. If the
observed effect is actually due to the hypothesized cause
- External validity -> If the hypothesized relationship also holds in other settings and other
groups (generalization)
- Construct validity -> Whether our methods actually reflect the properties we intended to
manipulate and measure
o Face validity -> That quality of an indicator that makes it seem a reasonable measure
of some variable.
o Criterion-related (predictive) validity -> The degree to which a measure relates to
some external criterion. The validity of a written driver’s test is determined, in this
sense, by the relationship between the scores people get on the test and their
subsequent driving records.
o Convergent/Discriminant validity -> Seeing whether the scores relate to similar and
different variables in a way that we expect.
An approach van be the multi-trait multi-method matrix approach: Use
different instruments to measure two traits. (Like cat fondness and pizza
fondness) -> conclude: no association between different constructs
measured with different methods.
You can see which measure instrument has low construct validity when the
measurements are not as expected.
o Content validity -> The degree to which a measure covers the range of meaning
included within a concept. (a math test needs to include the different parts)
Causal relationship
Hume’s criteria
1. Cause and effect are connected
2. Cause precedes the effect
3. Cause and effect occur together consistently
4. Alternative explanations can be ruled out
Introduction
Epistemology -> The science of knowing; systems of knowledge
Methodology -> The science of finding out; procedures for scientific investigation
Agreement reality -> Those things we “know” as part and parcel of the culture we share with those
around us
Errors:
, 1. Inaccurate observation
2. Overgeneralization
3. Selective observation
4. Illogical reasoning
a. Gambler’s fallacy (assuming a consistent run of either good or bad luck foreshadows
its opposite)
Three objections in regard to social regularities:
1. Some may seem trivial (“everyone knows that”)
2. Contradictory cases may be cited (indicating that the “regularity” isn’t totally regular)
a. An exception doesn’t always make a statement less true (woman earn less)
3. The people involved in the regularity could upset the whole thing if they wanted to
a. Does not happen enough for a serious threat
Variables of interest:
- Variables can cause other variables
- Independent variable: A variable with values that are not problematic in an analysis but are
taken as simply given. Presumed to cause or determine a dependent variable
- Dependent variable: A variable assumed to depend on or be caused by another.
Attributes -> Characteristics or qualities that describe an object
Construct -> Property in general, abstract
Variable -> Specific, concrete expression of the construct as measurable or manipulable.
Categorical variable -> Distinguish either nominal or ordinal categories . Does not make sense to
calculate a mean. Rather use the frequency or statistics.
- Binary/dichotomous variable: distinguishes between two categories.
- Polytomons: more categories
Quantitative variable -> Allow us to determine not only that people differ on a certain property, but
also to what extent they differ (for interval and ratio). It makes sense to calculate a mean
Discrete variable -> Can only take on a limited set of values. (Nominal and ordinal, yet quantitative
can also be discrete)
Continuous variable -> Always possible, in theory, to find a value between any other two values.
(between 6,5 and 6,4, there is 6,45)
3 Major purposes of social research:
- Exploration
- Description
- Explanation
Variables of disinterest
- Confounder/Lurking variable: Is related to both the independent and dependent variable
and partially, or even entirely, accounts for the relationship between these two. Is not
measured
