I’m starting my sociology studies to be better prepared for the redistribution research. I will share here my notes.
The name of the class is Methodological problems and construction of scientific-social knowledge.
Note that the class is in Spanish, and translating all the concepts is not super easy.
Knowledge as historical-cultural production
Research as historical practice
Subjective implication of human groups in research
Reflexivity
Complexity
Intentionality of knowledge
Plural approaches
Dialogue of knowledges
Complexification of language
Option for a viewpoint from the south
Transitions marked by multiple viewpoints:
Hermeneutic onthology
Understanding political action making visible unnamed elements from reality. Other ways of being.
Fenomenology
Recognize meanings given by the social actresses to their own subjective experiences.
Post-structuralism
Consider the games of power that condition and traverse the political agency of peoples and their subjective experiences. Recognize the shape-giving importance of material conditions and social forces in territory and life practices.
Decoloniality
From epistemologies of the south we can make decolonial practices of knowledge production. Situated perspective, not extractivist, critical, transforming, learning from resistance, recognizing the capacity of agency, and subjective displays to set up worlds.
Scientific knowledge is partial and conditioned. It includes, limits, omits and skews.
What’s interesting is the situated person.
Diversity of contexts, socio-historic determinants, topics of local interest, practice communities, commitments.
Plurality of epistemologies, paradigms, methods, disciplines, validity criteria.
It is a dialog.
Ontological rupture at the identity level: essential or dignitary, existential or biographic.
The cognizant subject does it situated and from the resources available in their context.
Those resources relate with who is being known and legitimize the produced knowledge.
Two views share the same process of knowledge, distinct and reciprocal:
In Latin America the first look from the alien other it the view of the conquistador. We keep seeing ourselves with their judgements and theories.
Recover the forbidden ways of knowing, with decolonization as prerequisite.
Epistemological rupture, revision, objection.
Formulation of proposals that admit plurality.
The diverse worlds coexist, they are superimposed.
Variety of causal explanations, immanent and transcendental.
Qualitative researchers see the world as people, events, situations, and the processes that link them.
The explanation lies on the analysis of the mutual influence between situations and events.
The research question originates on field observation, and is transformed during the research process.
The knowledge given by others is necessary and legitimate. It directs the selection of strategies for data collection and analysis.
Procedural validity:
Responsible researcher:
Triangulation, the term, refers to a concept that precedes it. It comes from geometry:
The concept comes to social sciences in the 19th century, before the debate of the quantitative and qualitative methods. The term appear in the 1960s, referring to independent measurements in quantitative research, which at the time was the only approach, following the scientific method.
Later it acquired new meanings: triangulation of indicators, researchers, theories and methods.
If the paradigms are antagonistic there can be no triangulation. In the 1980s the difference turns to a technical question, so a methodological pluralism is accepted. This brings validity to the triangulation and opens the door to coexistence, complementarity, articulation, integration and transcendence of methods.
Triangulation requires integration. If there is no integration, it is a research with mixed methods, which can follow parallel or sequential strategies. Triangulation is concurrent.
The metaphor of triangulation has been given a literal meaning. It has become trendy. It has been considered better than either quantitative or qualitative methods. It needs a critical and reflexive approach.
Looking for vertical coherence between paradigm, epistemology, methodology, method and technique is incommensurable, because their separate ontological origins.
Considering the relative autonomy of the techniques allows to resignify them, appropriate them in other contexts. It allows pragmatic articulation.
How to make observable the concepts of social sciences?
Social sciences are not factual sciences.
They are oriented to give an account of the world. To generate theories that allow to produce knowledge with respect to the evolution of social phenomena.
Most of the social concepts are not directly observable.
To contextualize the concepts in their corresponding theory. It is hard to integrate one same word between theories. E.g. social class, marginalization.
Find empirical indicators related to the unobservable concept.
E.g. poverty → how they dress, where they live, what kind of house they have, what kind of education they have.
The problem of epistemic correlation: correlation is unobservable.
Literary description of the concept.
Elucidation, theoretical work.
Identification of the concept dimensions.
Choosing the indicators
For each dimension, one or more indicators.
Building the indexes
Synthesis of the indicators, observable representation of the concept that we want to measure.
One part is directly related, plus one part of inherent error.
How big is the error?
In qualitative research, use the saturation method. Repeated observations over a same individual.
Are we measuring what we want to measure?
From Real Academia Española:
It assumes that we have direct access to reality. Let’s leave this aside to see the compromises that we can reach in a survey.
From Wikipedia:
It is usually associated with the application of a questionnaire, but it can be done y many other ways.
We won’t call it representative sample, we’ll call it probabilistic sample.
Questionnaire: list of questions that in social sciences have the epistemic goal, to build knowledge.
Epistemology: theory of foundations and methods of knowledge.
In this class we will call survey a printed questionnaire with structured questions applied to a probabilistic sample to register information, as in instrument of measurement and concretion of a theory.
Generally they are applied face to face to people, or in front of documents. They can also be self-applied or by phone.
Units of observation have to be decided. Units of registry are the units of observation and others units included in them. Units of analysis are constructed from the units of registry.
Strategies to design research. Methodological pluralism uses them together as complementary cross-control elements.
Comparison: controlled observation of different phenomena to see their variations and similarities.
Social phenomena are subject to certain regularities, that explain them and predict what is going to happen. It’s a paradigm of scientific reality, not positivist.
The comparative method tries to establish regularities between observations, to generalize situations and causal explanations.
It involves statistics, many cases.
Phenomenological tradition: social phenomena are singular, they have situated historical content that makes them unrepeatable.
It doesn’t make sense to understand the regularities. We have to interpret the fact in its context.
Understanding from the particular complexities. More inductive. From the specificity of facts, understand social phenomena.
Systematic study of a determined phenomenon situated in time and space.
The limits of the specificities are defined in function of the scientific, personal, social and political interest of the researcher.
Once contained, try to understand the depth and complexity that makes emerge the singularity of that phenomenon. How it is produced by its interaction with context, leads us to its comprehension.
Sometimes there is no clear boundary between context and phenomenon. This is its complexity, diversity and richness, that has to be shown in the conclusion of the study.
Picking the case defines the object of study and it’s methodological tooling.
Options:
Atheorical case study. Pick the case based on its importance. It’s more ethnographic, more phenomenological.
Interpretative case study. Keep the importance of the case, and try to understand it with a theory already defined. Does not touch all the aspects of the case, just the ones that the theory allows to observe.
Test if a theory is valid or not in the context of a particular case.
Crucial or archetypical cases. They are so particular that they have a lot of value in the verification of a theory. Used to put a theory at risk.
Compare more than one case. Try to analyse similarities and differences between phenomena.
Select units and properties to compare. Do not go deep in all the complexity to understand emergence. Make comparisons trying to control the observations. Establish certain regularities to generalize.
It’s more hypothetical-deductive.
Questions:
The researcher as a knowledge tool.
Knowledge practice that comprehends social phenomena from the perspective of the protagonists.
Texts of monograph and visual presentations in which the researcher represents, translates, interprets a culture, a social group or some aspect of that culture that is presented as exotic. There is some estrangement.
Method: field work
The same person goes from start to finish.
Amassed between data and theory takes shape of description, to advance in a science that is getting less eurocentric, to find concepts really universal.
Continued, thorough, in person, holistic. The field impacts the researcher, takes them inside. The conclusions come from a real articulated relation with the population that we want to know.
It is an internalized interlocution that is produced using the researcher as a tool of knowledge.
Participatory observation
Interview
It is in the tension of the relation where knowledge appear.
There is no unity, universe, or place, because the destination is not known. There is no method, because the way they know is unknown. This gets resolved with an exploratory trip.
Preserve the flow of the daily interactions in the community, not of the academic chore. The investigative tools are the ones that are available.
A property of social life. The activities done to produce and manage situations in social life are the same as the procedures that we use to describe these activities.
Referential and performative value of the language. It takes time to understand that both researcher and observed generate a same common world.
Control reflexivity seeing it from a critical perspective, understanding how we are seen so there is communication, comprehension, understanding and knowledge.
Method to obtain information based on interlocution, question/answer or stimulus/response. It is an artificial situation in which interaction is mainly verbal.
Seemingly natural interaction, spontaneous and free. Actually, it is a controlled situation, systematic and with professionally defined objectives.
The interviewer is not a neutral collector of information.
It is a co-construction between interviewer and interviewee.
The data obtained are the result of a unique and unrepeatable situation.
Depth and detail. Low budget.
Conversation group. A device for the knowledge of common places.
It can be defined as a scene in which 6 or 7 participants, who don’t know each other, talk, guided by a professional person, about a topic that is relevant to their daily life.
They come from the same social macroset, which they represent.
The result is a consensus, the sense that unites them in a group.
It is a verbal replayer of the prototypical discourse, a text of the common places, of the word of the group.
The place: a plain in the woods where a human group reunites, with the same name, closed.
Kant talked about the rational conversation.
The qualitative method promised to listen a voice that was not know. In 1983 in Chile, where was the people? We had to start from listening. The liberal subject disappeared and the socialist, and the formalist, and the civilist. They were trying to build the neoliberal subject. There was no god, no polis, no class.
The group promised to know this new subject wherever it was, amongst peers, when they were capable of common word.
Take the group to arrive at their places, there where subjectivity appears weaved with the other, like a speech and its consent integrated.
A practice of qualitative observation, reproducing group conversation of social topics.
Compared to interview: it is a group speech, not talking to the other. Both reproduce a text of common sense. It’s a common place, not a self place. Internal coherence or consensus within the community that talks.
What is the norm of the group? The most sayable, listenable, consensual. A shared listening that looks for consensus. They rationalize to select the thinkfeel in community.
The subject in front of the group, in front of the other, talks appearing, showing their adjustment or maladjustment with the collective word.
Two articulated dynamics:
Roll in a set, unroll in a text. To group, to text. When talking, propose a possible place, weaving according to the sense of what was said before.
They talk developing knowledge and perception that is folded in their field of obviousness. They consider it significant in common.
Test of group consensus: the group becomes strong when the consensus is intensive and extensive. The group disappears when the conversations don’t flow.
The text is what remained. The thread vanishes and they return to their separation.
Procure that the map of common places is reproduced, that the entire discursive field that opens up is traveled.
It most happen from inside.
From what is heard, everything is pertinent.
From every element that appears, you can jump to the sides, as a metaphor, backwards of forwards, as storytelling. Following clues that the participants give as an answer to their previous stimuli.
The rhizomatic guidance, risk not having a map, asymmetric architecture, decentralized. Follow a trace, the group appears when it is followed. Incitements that lead to talking.
Representative information from a few meetings of few participants not randomly chosen. No one is an individual, they represent a version of their complete structure.
The qualitative sample is representative if its group has the same form and structural composition of its universe.
To study the impact of the course of time in the biography of people.
It can be about the biographic facts of a life, or the experience of the people through discourse or biographic memory.
Focuses on the presence of individuals in the social life.
E.g. the trajectory of those who start attending a university. With their id number they are followed, the facts are their first subjects and exams.
Doesn’t need facts, it pays attention to the reality of the language. To the form people use to tell their life story.
Not always real stories and life facts can be told. A life narration is an interpretative achievement that includes many contexts and is very valuable to social sciences.
1918: The Polish Peasant in Europe and America. The book that uses letters for the first time as the surface of sociological observation and a life story.
1923: The Hobo. A sociology of the homeless man. The author interviewed and collected stories from many people.
1930: The Jack-Roller. A boy with issues with law enforcement, institutionalized several times. Opens the window to the world of institutional lock down and brings new visions and variables to the sociology of crime.
1961: Los hijos de Sánchez. Collective biography of a poor family in México City. It can’t be that the impacts of modernization of the Mexican economy are dramas only for novelists and social reformers. They also have to be the object of sociology and anthropology. Criticized for being too raw, it broke the image of the golden age of Mexican cinema.
1963: Stigma. Based on testimonies, not facts.
1980: Biography and Society, The Life Story Approach in the Social Sciences. Product of a study group with Daniel Bertaux.
There is no common terminology. It’s more interesting to see beyond the words, what does the researcher do?
Reconstruction of sociostructural entities. Through the testimony of actors or follow up of their individual behaviors. Statistics are useful.
Reconstruction of group cultures. An episode marks before and after in the life of individuals, and starts processes of desocialization and resocialization, relational and cognitive. There is a biographic break or turning point.
Realization of microhistory. A moment with recognized historical significance is recreated through the testimony of actors that lived it as witnesses or protagonists. Reveals non-official histories.
Revelation of narrative marks. The narratives of the self are discursive constructs, with them individuals experiment social identity. Life is recreated in the story with language marks.
To study people’s life stories has the same analytical value as studying the form that people use to tell their life story. Facts or discourse.
Credit card payments, GPS and maps, internet of things, social networks…
Digital fingerprint: collecting information without the subject losing the focus on their task.
Data deluge: so much data that we don’t know how to handle it.
A new paradigm. Techniques and technologies are a new kind of scientific exploration.
Are surveys over?
Machine learning. Before, the model came from outside, and data was used to estimate it. With a lot of data we can construct, estimate, and re-evaluate the model as it is being used. Most or all of the model construction can be done by a computer.
Autonomous driving, smart cities, economic research on inflation based on retailer prices, credit scoring, predictive surveillance, insurance pricing, facial recognition.
Companies buy the data we leave in the internet to execute their analysis.
Facebook can predict a lot identifiable characteristics from a small sample of likes. Gerrymandering, modified the cartography to benefit its party, concentrating their votes and spreading the opponent’s in 1812. Alexander Nix, from Cambridge Analytica, behind Trump’s and Brexit’s victories, a political campaign based on the preferences of the voters.
Enough of the human behavior theory from linguistics and sociology, forget taxonomy, ontology and psychology. Track what people does with precision, it doesn’t matter why they do it. – Anderson, from Wired.
We often don’t care about causality, the goal is more prediction than understanding, it just needs to work. – Singer.
Big data sets allow data to speak. Using a random sample is like using a horse instead of a car. Valid hypothesis are no longer necessary. Knowing without understanding.
Not well defined coverage. Sociological regression. Impossible to situate the producers of numeric opinions. Individual as a mine, an archive. The researcher has to do data mining, extractivism.
Hawthorne effect, the observer can change the behavior of the subject.
Sentiment analysis. Analyzing social networks requires only to listen to the flow of unsolicited opinions. Non mediated by an observer, unobtrusive public opinion. It’s reality, non reactive, objective.
The simple act of creating a profile in a social network requires reflection. Think how they want to be represented.
Spiral of silence. People are afraid of being isolated and of social sanctions. They tend not to voice their convictions that are in contrast to general opinion.
Social desirability. Culturally and historically situated. Subjects decide what is the acceptable answer and image, often without knowing.
Self-deception, to deceive or give self-trust, to strengthen self-image.
Goffman, masks and facades.
Digital fingerprints doesn’t sound accurate. Graffiti sounds more accurate.
Big data as cheaper? requires less human researchers. Buying data is expensive, firehose. Small samples might be free, garden-hose.
Differences between rich and poor universities.
Can big data be the foundation of the society of knowledge?
Who owns the information?
Companies decide what can be researched and who can do it.
There is risk. Universities and independent researchers are being excluded. A dualistic system, scientists working for big companies will enjoy all the data, researchers not in silicon valley will only get fragments. Computational data science might be dominated by private companies, government agencies, and privileged academic researchers, that produce knowledge that can’t be verified or disproved.
Deluge or desert?
Privacy? Erroneous conclusions if data are widely published?
Ethics: interested in the judgement that orients the moral action. To form an ethics code in research implies to identify the values and principles that guide the researcher.
Just bringing this topic up accepts the lack of neutrality.
Biomedic research has established a series of ethical codes. Research on human beings.
Social research doesn’t have a set of norms, might not be able to have it. Research with human beings.
We can’t always reach the clear consent.
The ethics depend on the trust and accountability of the researcher.
Sociology cares more about the external expression of humans in society.
Social psychology is a grey area that also cares about the internal expression.
Relation with the subject and relation with the public. Researcher has to account for their point of view.
The view is different from the worker, employer and government. There is no single truth.
We are trying to build a science. It is collectively constructed by a group of researchers.
General ethic norms of science.
Absolutist. Only research public domain phenomena.
Pure scientificism. What matters is the truth. Consequences are not important. Requires deception.
Relativism. Ethical rules have to adapt to their context.
Feminist. Search for the personal accountability of the researcher, empathy with the subjects, share emotions, value the expression of feelings.