Comprehensive Guide to Descriptive and Observational Research Designs in Psychology

Hello and welcome to the course basics of experimental design for cognitive psychologists. Uh this is week two and

uh before we get into the details of the experimental research design uh I will present a brief survey of the uh

descriptive and uh correlational research designs as well so that we have a good idea about what are the strengths

and what are the weaknesses of the experimental research design and also uh that we are aware of the uh you know

considerations in different uh other research designs. uh basically to help us be a to help us decide what kind of

research designs will suit our research questions. Today I'll talk about descriptive

research designs. Let's proceed. Now uh we've seen uh you know uh that there can be different kinds of research

methodologies. There can be quantitative, qualitative and the assumptions are different and the

advantages and disadvantages are different. So let's let's see uh you know let's dive a little bit deeper and

try to understand uh one of the most important research designs that is the descriptive research design. Now

naturalistic research naturalistic research designs is a type of research design or is it a is one of the

instances of descriptive research design that are designed to uh describe and measure the behavior of uh the people or

animals as it occurs uh in natural settings. So when you're talking about uh naturalistic research then you're

basically uh trying to collect data gather data as it occurs. If you are interested in uh studying some kinds of

human behavior, if you are interested in studying some kinds of phenomena unfold, uh the idea in naturalistic research is

to not be a party and not manipulate or control for anything but just let the behavior occur in its most natural form

and be able to observe and collect data in that particular method. Now, uh how would you collect data in such a uh

scenario? So ways of recording behavior in such a research design could be for example that you make recordings uh as

it occurs or it could already have been recorded by others. Say for example uh you know you can uh pull out CCTV

footage at a busy intersection and uh focus at the turning behavior of individuals. Say for example, do

individuals necessary give uh you know an indicator when they are making the right turn or say for example are they

speeding up or speeding uh down when they reach an intersection and many things uh you know like that.

So uh either you sort of you know make arrangements to record behavior uh at uh you know an intersection of choice at

you know with a particular kind of target population in mind and so on. Or you could just go back and see okay uh I

want to study the uh you know the turning behavior or the traffic behavior of individuals. I will basically you

know uh pick out data from these different intersections. So you can choose the intersections and so on. All

right. Now uh the data that would be obtained from such uh you know recordings uh for example uh would be

rather rich. It would be rather complex and it will need to be organized into meaningful uh measured variables. so

that it can be analyzed and interpreted. For example, even let's let's stick with this traffic uh example for a bit. Now,

uh there will be different kinds of vehicles. There will be say for example people on uh foot, people on cycles, uh

people on scooters, mobiles and so on and people with cars or heavy trucks for example. You will basically be able to

collect data from all of these and uh it is possible that the driving behavior or let's say the turning behavior uh to be

more specific in this case might depend upon the kind of vehicle the person is uh riding. It might depend say for

example on the time of the day. It might depend on the uh you know amount of traffic there is at the time uh you know

the recording is being done or the time from which the data is being picked up. So uh basically what you'll see is that

there is a lot of data. It is rich. It is complex. It has all the details. Maybe there are particular uh you know

intentions, particular times uh that uh govern people's traffic behavior and so on. So uh if you want to make sense of

this data, one of the first things that uh you know this researcher will need to do is to basically analyze and

categorize this data into measured variables. Remember we've talked about operational definitions and so on in the

previous lectures. So you'll need to create operational definitions. You'll need to specify the variables you are

interested in and that should basically govern how you go forward with you know categorizing, analyzing and later

interpreting this data. These designs the naturalistic research designs are have been actually extremely popular

across a variety of disciplines which have been interested in observing the you know natural behavior of

individuals. For example, these are obviously interesting uh for psychologists but also for sociologists

also for people who studying media and anthropologists for example uh you know geologists or say for example animal

conservationists and so on. So this particular technique is one of the most widely used techniques and it is used

has been used across a variety of disciplines and the data that forms the basis of this naturalistic research

method uh can also be uh consequently drawn from a variety of different uh methods. Let's talk a little bit about

that. Now for instance uh in some cases these data may be derived from a clinical psychologist's informal

observations about his or her patients or other scientists more formal observations of the behavior of other

animals in the wild. So for example when you visit a doctor when you visit let's say a clinical psychologist for that

matter and you're having an interview and the psychologist is asking you a bunch of questions uh there are

obviously the questions that the psychologist is putting to you but he's also observing say for example uh

various behavioral takes how comfortable or easy are you on certain topics how uh upcoming uh you are in terms of uh you

know the kind of answers that you're providing how detailed how shallow uh your body language your emotional state

and so on. So all of this basically is also data that the psychologist is using. Similarly, for example, I'm sure

uh you know um you might have come across uh you know documentaries from Discovery Channel, National Geographic,

Animal Planet. There are so many of these uh you know uh which are available and basically what they're doing is they

uh put the camera somewhere and they go back and they're basically observing how this uh you know natural behavior, how

the animals are behaving in their natural habitat without uh you know controlling for anything. And later by

the way whatever you see is basically uh you know these observations which have probably occurred over a large amount of

time uh at different places being knit together uh using a narrative and that's why when you see it seems like oh it's

it looks like almost scripted animal behavior. All right. Uh also for example uh you know if a developmental

psychologist is uh interested in studying uh children's uh play behavior or children's behavior with their

parents. So they might basically videotape uh children playing with their parents in a laboratory setting. So in

some cases you can have a relatively control setting. You're not interfering with the behavior but you say for

example have room where the parents are there, the children are there, there is a lot of toys. uh you're observing from

the other side of a one-sided mirror and you are making this whole whole video recording and later you will go back to

this data. You'll organize this data. You'll uh try and analyze according to certain predefined categories and then

uh you know uh interpret uh basically what the data uh you know really says. Now uh predominantly the naturalistic

research design as we've already said you know it yields rich and complex data. So predominantly uh these

approaches lead to uh you know qualitative data. So uh remember the things we were talking about in the

previous lectures. So you have to uh analyze them, code them uh you know arrange them according to themes and so

on. So predominantly it'll lead to qualitative data but some cases you can also code this data in form of

quantitative data as well. And we'll talk uh a little bit about that as we go forward.

Also sometimes naturalistic uh research is uh adopted because it is the only feasible uh approach of gathering data

and studying a particular phenomena. For example, uh researchers may not be able to study the impact of say for example

earthquakes or flood or cult memberships using experimental designs. So uh once the earthquake has come uh and people

are behaving in a particular manner, they're trying to save themselves. They are affected by it emotionally as well.

uh there's a lot of uh stress involved and so on. You cannot really replicate this situation or you cannot basically

you know make comparisons across situations across individuals and so on. So these things cannot be studied using

experimental research design and therefore uh a variety of uh you know uh uh data can be basically collected when

you're following a naturalistic research design. Obviously uh you know in in much less uh exciting uh situations for

example uh when people are affected by calamity or terminal illnesses and so on naturalistic observations might uh you

know help uh in getting a sense of uh how their behavior is. Also uh a particularly important concern uh with

naturalistic research designs is the fact that it highlights the importance of what we call ecological validity.

Remember we were talking about uh you know one of the possible drawbacks with experimental research because everything

is so sanitized everything is so controlled uh and you're basically uh instigating or eliciting certain types

of behavior in controlled laboratory settings. Uh it ceases to be the same behavior that you initially wanted to

study. Uh you know you wanted to uh you wanted to probably study something uh like how do people remember information

but in a general classroom setting. uh but because you wanted to follow the experimental research design you control

for X Y and Z but then what happened is that when you were conducting these experiments in the laboratory with very

fixed lists uh with a timed exposure to particular information and a time pressure in eliciting the responses

you've basically overs sanitized you know for some or sanitize the environment and thereby what you've done

is you've reduced the ecological validity what is ecological validity it is the extent to which uh the research

conducted in situations similar to the everyday life experiences of uh you know participants. So uh what happens is when

the ecological validity is down the generaliz generalizability and the interpretation and the validity of that

behavior to explain phenomena happening in general circumstances uh basically gets reduced a little bit. Okay. So the

idea here or the one of the fundamental ideas of naturalistic research designs is basically to capture the behavior of

people as it would occur in their everyday life uncorrupted by the influences of the observers or the

artificial setting. Uh and say for example in some sense the researchers would try to ensure that you know there

is minimal reactivity. the participants are not reacting to the situation or the experimentter and therefore what happens

is that the construct validity and the ecological validity is relatively maximized. It's sort of optimized if you

might want to use that word. Now let's look at some other types of research. So we've talked about uh naturalistic

research. Uh there is also uh you know a similar uh uh thing which is referred to as the observational research design.

when you're making these observations a very interesting case is uh you know is presented in this book. For example,

Roy and uh Roy you know uh this British psychologist in 1959 to 960 wanted to observe the behavior of workers in a

raincoat factory. So what he does is he takes a job in the factory which entailed uh you know where these

raincoats are being made and the kind of or the nature of the job is basically rather boring. So uh this uh job entails

repetitive uh movements basically punching holes in a plastic sheet to say to sort of uh you know uh cause make

buttons and so on and this sort of went on for 8 hours a day 5 days a week. Now Roy was not really sure uh about how the

employees some of whom uh who had been working there for 10 20 15 you know 10 15 20 years could actually stand the

monotony and the you know and the boring schedule of just uh you know uh taking plastic sheets punching them maybe some

others were involved in cutting them and so on. So he thought that this is an extremely boring task and it might be

interesting to study as to how do the participants react to such a boring environment.

So in the first few days on the job, Roy did not really notice anything particularly unusual. However, as he

carefully observed the activities of the workers and uh you know workers or the employees over time, he began to

discover that there were ways in which these employees were actually keeping themselves entertained, keeping the

energy sort of variable uh in in the work environment. For example, he discovered that these employees had a

series of pranks up their sleeve that they played on each other and in some sense it kept the uh you know it kept

the environment light. It kept them interested and uh awake uh in basically executing these work schedules. For

example, there's this individual called Sammy. Uh every time he goes to the drinking fountain, uh Ike, another

worker, turned off the power on Sammi's machine. And when Sammy returned, he tried to stamp a piece before

discovering that the power was turned off and would basically discover that oh, the power is not working. And he

would be angry because he he knew that somebody has sort of done this. So when he uh acted angrily towards uh you know

this gentleman Ike who in turn responded and basically said oh I don't know and he basically shrugged and smirked. So it

basically creates a bit of a you know a fun a banter sort of a movement. Also in addition to this kind of event

uh something all uh you know something else also happened several times today. For example, Roy noted that uh there

were several other games that the workers effectively effectively used to break up the day and keep up the

excitement. For example, at 11:00, uh Sammy would yell banana time and steal the banana of Ike's lunch pail, which

was kept somewhere there on a shelf. And uh what would happen is later in the morning, Ike would open the window in in

front of Sami's machine and basically, you know, let in the freezing cold air. Sammy would go protest and close the

window. Uh also at the end of the day for example Sami would quote two minutes early just two minutes early to sort of

uh you know uh just instigate people around him drawing fire from the employee boss uh who nevertheless

anyways realize that this is harmless and it's not really a lot of uh you know uh

loss of work and he basically let this activity occur throughout the day. So you can see there are a bunch of things

that these uh you know participants these workers are uh you know employing they are utilizing throughout the day to

break the monotonous experience to break this uh you know boring kind of day that they they would otherwise have if

everybody kept to themselves did not interact and sort of behaved in a particular manner. Now uh although Roy

entered the factory expecting to find only a limited set of mundane observations, he came back with a whole

uh you know he discovered a whole world of regular complicated and uh you know to the employees satisfying activities

uh that broke up the monotony of their everyday work existence. And in that sense he discovered that okay there are

these uh reasons there are these activities that keep up the morale keep up the uh you know excitement of the

employees keep keep them engaged and that is how basically they sort of were able to survive in that environment day

after day. Now uh there is a bit of a caveat here that this represents one of the and

which represents one of the major advantages of naturalistic research methods because the data that Roya was

able to collect was rich. It was complicated and it could also lead to several new research ideas. Say for

example, if fun activities are not allowed if uh you know the timings are sort of shuffled uh what would happen

say for example if uh the task did not uh require a lot of manual effort and it sort of become became too automatic and

so on. So you can basically uh on the basis of studying that environment uh come up with several hypothesis and then

if you want to see you can choose different settings where uh one of the variables is slightly different and

study okay how do these people work uh out their day and how do they sort of uh uh you know survive throughout their uh

you know work lives. Now we can see however that as the researcher was working a stamping machine and

interacting with other employees, Roy was a participant in this setting. All right? So he's not outside the setting

and observing from uh you know outside and basically you know a bird's eye view. He's actually involved in the

setting. Okay. uh so when a scientist uh like this or in in this setting when a scientist takes a job in a factory uh

let's say joins a religious cult to study the behavior of the other members of the religious cult or let's say even

sometimes you know checks into a mental institution he becomes part of the setting itself.

Now this uh may have advantages may have even disadvantages because it can probably influence the behavior of the

people around. So let's let's talk a little bit about that uh going forward. So this is one uh kind of uh you know uh

way in which data is collected. Other times what might happen is that the scientist may just choose to remain

strictly an observer of the setting such as uh you know when he or she views the children in a classroom from a corner

without playing with them. You know a lot of times uh uh I don't know if you remember in school uh somebody would

come for an inspection and they'll basically quietly enter say sometimes the principal and the vice principles or

the you know head you know uh head instructor for a particular course might uh quietly slip uh you know from behind

slip behind the classroom and they'll basically study how a particular teacher is teaching how the kids are interacting

and so on. So in in these cases you see that what is happening is that the researchers or people who are interested

in collecting this data are choosing not to become integrated in the setting. They're not becoming part of the setting

but just observing the natural behavior in the setting from a distance. Also in addition to deciding uh whether

to be a participant the researcher must also decide whether to let the people know you know that the observation is

occurring. Say for example if uh there is an individual let's say there is a individual uh who has gone to uh let's

say uh observe a particular classroom interaction okay so there is a new teacher being hired uh and one of the

senior teachers basically say oh I just want to understand how this new person is teaching and how are the children uh

going to interact with this teacher and so on now uh in those cases you'll basically see that the uh senior teacher

might tell and inform everybody that okay a particular ular kind of observation is occurring. Now this is

basically what is referred to as an acknowledged uh you know study or acknowledged observation in some cases

where the uh you know uh data collector or the researchers surreptitiously mingles with the environment does not

tell anyone that they are observing uh anything a phenomena or an individual or particular kinds of behavior then that

would be referred to as unacknowledged observations. All right so we'll talk about these things also as we go ahead

in some detail. Now as the decision about whether to be participant or non-participant and say for example

whether the observation has to be acknowledged or unacknowledged typically it leads to four kinds of observational

research designs. We'll very quickly uh go through them in some detail. Let's say now this is basically the broad uh

thing. Uh so unacknowledged participant we have the Royy's example. He was unacknowledged. He did not tell people

that he was observing. acknowledge participants. There's this uh you know we'll talk about Weiss's study of street

corner society. So White basically wanted to study a particular kind of a society and he went there and he told

people that okay I'm observing uh you and I'm making some notes. Unacknowledged observer recording

behaviors of people in a small town uh acknowledged observer say for example prominence in colleague study of

children's social comparison habits. We'll do all of this in detail and let's let's see uh how it goes. Now uh first

let's talk about the unacknowledged participant. Now this is the same approach that we saw uh you know playing

out in Royy's example. Now uh while it allows for an advantage in observations because you're concealed and you're

basically uh you know uh not letting people know that you are observing anything and in that sense it will still

uh most probably elicit natural uncorrupted behavior. It'll uh the reactivity is minimized. the people are

not conscious and they're behaving as they would normally do. Now, uh while so this is while is interesting and it lets

people reveal their truths, levels and so on. But in cases where the data being collected is about personal information

or intimate information and social situation and opinions and attitudes of people then basically uh you know it

might create problems for the people being observed and in that sense be unethical.

All right. So it could be possible that the observer might sometimes find it difficult also uh to remain objective.

So when the per when a person is uh you know uh participating in the setting and has become part of the setting uh they

are also affected by whatever is going on they are also probably affected by the monotony or the two or the

excitement of the situation. They are affected by people's behavior, their attitudes, what they are saying, how

they are treating each other and because the person is in the setting and everything that transpires in the group

transpires through this person as well. So as a result, what might happen is that the person may become too involved

in the situation or with the people and find that uh their opinions and perceptions are getting affected by

their point of view. So the researcher bias might come in. the researcher likes a particular person, does not like a

particular person, has a certain opinion about the overall you know scenario uh positive or negative and that will start

affecting how they transcribe and how they later interpret and analyze the data. All right. So all of this

basically makes these designs slightly more difficult to execute or at least present certain challenges. Now the

other concerns uh you know that is uh raised uh with this kind of research design is that it poses ethical dilemmas

for the researcher. Something I was just mentioning. For instance, people being observed may never be told that they

were part of a research project or may find out only much later and in that sense they might have revealed things

that they did not want to uh put out in the public or make available for uh you know uh analysis and research. See one

of the fundamental aspects about ethics and we'll talk about uh you know ethical consu considerations in research at some

point but one of the fundamental thing is that the data must be taken with some kind of consent. Uh if you are

collecting data uh you know in in a way that the participants being observed are not uh knowing that then also you should

uh the researcher should find a way to debrief the participants at at some point. All right. So uh this uh

basically this kind of a situation where the participant is unacknowledged and the participant has not really let

people know that they are making notes they are observing and they are sort of you know creating this uh data pool. Uh

this would be a problem when the observations say for example start including personal details intimate

information. uh also identifying information. Say for example, if Roy were making these observations and one

of the employees made a snide comment about the boss or about the workplace uh and later this information is

identifiable uh then that person might lose his or her job. So in that sense you can see probably the uh ethical

considerations that might come in. Finally uh you know another disadvantage of the design is that the activities of

the observer may start influencing the uh processes in the phenomena being observed. For instance Roy uh if say for

example he is going through a day he's also working on a stamping machine for example and Sammy and Ike and others are

there now if for example a group discussion uh uh you know uh erupts people are talking about something

they're talking about a certain uh employee or the boss or the overall situation. Now if he would say nothing

and not contribute it would probably blow up his cover that he's doing something else there something fishy is

there or if he says something interesting or something substantive whatever he has said might shape the

discussion might shape the nature of the group interaction and that again you can see corrupts the kind of data uh the

person is getting and the person is making record of. So uh another possibility is that the

researcher becomes what is called the acknowledged participant. So we're talking about the unacnowledged

participant. Now we are talking about the acknowledged participant. Okay. Sometimes when the researchers feel that

it is unethical or impossible to hide his or her own identity, they may use the acknowledged participant approach.

For example, sociologist White uh used this approach in a sociological study of the street corner society. And what

happened was that over a period of year, White got to know the people living around this location,

made extensive observations of the neighborhood in this New England town. Now, he did not hide his identity,

rather announce freely that he was a scientist and he wanted to just make notes and observe how people uh you know

live and behave in in this kind of a society. Now sometimes this approach is uh extremely important because when the

behavior that the researcher wants to observe is difficult to gain access to say for example he wants he wants to be

party uh to uh let's say certain you know colony meetings or society meetings uh if he's not a member of the society

uh he will not be able to uh gain access to that and if he is a member of the society but he does not disclose that he

is collecting data then also that may be unethical and uh you later may carry legal etc implications legal legal and

ethical implications. So also say for example if you want to uh record a boardroom conversation a particular

meeting and so on uh you will not gain access to if you uh are not acknowledged uh if you're not a participant and also

if you not really acknowledge the real purpose of your presence. Now uh the most important problem with this uh kind

of research design the acknowledged participant is that of reactivity. The knowledge that a certain observer is

recording information actually causes or can cause people to change their actual behavior and cause them to pick and

choose their topics or utterances or to avoid the researcher altogether. So either you'll not get any data or you'll

get mostly acting data. For example, I'm sure everybody is familiar with the uh you know with the series Big Boss or Big

Brother. Now everybody is aware that there are any number of cameras in the house and uh while you'll see that the

participants uh quite often try to sneak into places where the cameras are not but for the most part because they are

so heavily surrounded by the cameras they are behaving in a uh you know in a particular I would say maybe artificial

maybe not uh way that they're at least conscious of that there are cameras and recordings are happening. Now if this

goes on over time uh research suggests that once the observer say for example has spent enough time with this

population of interest or let's say if the cameras have been uh are around throughout the day uh 24/7 for several

days people may sort of get accustomed and normalized with this whole thing. Okay. So people shall begin to treat

such an observer as a real member of the group. They might start having conversations. they might sh start

sharing details with this individual uh without really caring for whatever this guy is observing or not observing. So

this is probably something that happened with the sociologist WF White uh as he allowed the people in the observed

society to get habituated his friends uh habituated to his presence and started his recordings uh you know only

afterwards. So he took care of the fact that the observations are not uh you know eliciting reactive sort of

behavior. Now nonparticipant observers. So

acknowledged and non-agnowledged. Nonparticipant observers. Let's talk about this. So sometimes it is possible

that a researcher may uh you know utilize a nonparticipant approach when he or she does not want to or cannot be

the participant in the kind of group that is being studied. In such cases, what would happen is that the researcher

will try to observe the behavior of interest without actively participating uh in the ongoing action. For example,

when children are observed in a classroom, I was talking about this uh from behind a one-way mirror or when

clinical psychologists are trying to videotape group therapy session for lateral analysis. So the individual, the

researcher is away. He's not participating in the action but he's still making notes and keenly observing

all the details. An advantage of this kind of an approach is that the researcher can be extremely objective as

he or she is not participating in the situation, is not developing relationships with the people being

observed, is not developing opinions or uh you know and is therefore being able to more critically evaluate the

phenomena that is unfolding in front of their eyes. These nonparticipant observers can be either acknowledged or

unacnowledged does not really matter. But we've seen that this has its own pros and cons. So when you are uh told

people that okay I'm observing from a distance or you've told people that uh you've not told people that I'll be

making these observations both as we've seen uh acknowledgement and non and non-agnowledgement basically have their

own uh you know uh outcomes. So I'll stop here uh in the next lecture I'll talk to you about uh another type of

research design. Thank you.