Naturalistic Observation Paper
Introduction The purpose of this study was to conduct observations, using two different methods, on children, between the ages of 4-5, to understand their emotional behaviors in a social and learning environment as well as determining which method of observation was more reliable. There are different methods of conducting observations, but two main ones include naturalistic observation and experimental observation. A naturalistic observation method is when a researcher studies a person or group in their natural environment or setting; the behavior that is observed is one that occurs in a person’s everyday life. In contrast, an experimental observation method is when a researcher is studying a specific behavior in a laboratory-like environment.
…show more content…
This type of method observes the cause and effect of a behavior by making the behavior the dependent variable and having a control group, a group that doesn’t receive the change to produce the targeted behavior, and an experimental group, a group that does receive the change, and measure the effect of the change. The benefit of the naturalistic method is that, unlike the experimental method, the results show a more direct relation to a behavior and reality since experiments are conducted in an artificial standpoint, and more than one behavior can be observed. However, the experimental method allows researchers to study a specific behavior by isolating it, which also helps them to be able to control the situation and record more information, rather that losing it because the researcher was recording a non-targeted behavior. In this specific study that was conducted, the two naturalistic observation methods that were used included running record global observation and time sampling with predetermined codes. The running record method is when researchers records everything that happens including behaviors that are the main focus of the study. For an example, a researcher might be observing a child and wants to know their interactions with peers but they would also record interactions with teachers and other adults if it were to occur during the observation period. The time sampling method is when a researcher observes a desired behavior in set time intervals. The behavior is dissected into small actions related to the behavior, and are put into a coding system. For an example, if a researcher is observing how a child interactions with their peers, they would record actions related to this behavior, in small intervals-i.e. 30 seconds- and use codes to describe the behaviors, i.e. playing with peer=1, talking to peer=2, not interacting=3, etc. When looking for patterns in children’s overall social behaviors, using the running record method would be advantageous because it can help quantify the frequency time-wise, of the behaviors, thus helping to analyze how often a behavior occurs and how soon after another behavior occurs in response to the original behavior. The time sampling method would also be advantageous because it can help show the pattern of the behaviors interacting with each other in a sequence. The disadvantage of using the running record method is that the researcher can be overwhelmed by all of the behaviors that has to be captured and consequently misses information; for the time sampling method, the researcher can lose information showing the flow and actual relation between each behavior. If the researcher was only observing one specific behavior, the running record method would be advantageous because it could, again, show the time frequency of a behavior; likewise, the time sampling method could assist in quantifying the frequency of how often a behavior occurs in allotted intervals. However, the disadvantage with using the running record method is that the researcher would be losing extra information that can assist with learning more about a behavior. Also, with the time sampling method, the researcher would still lose information about the time of the occurrence of the behavior. The naturalistic observation that should give the highest level of inter-observer agreement should be the time sampling method. This assumption is due to the fact that this method has an agreed time frame; both of the partners are recording at the same time. Also, since we are looking at the same behavior with a code system, there shouldn’t be any discrepancies over the recorded behavior, and the coding system allows the researcher to be able to quickly take down more observations.
Method
The procedure for the running record observation method my partner and I used, was that we sat in an area where we were able to see the children but they were not able to see us in their field of vision. Then, once settled, we observed the child that was most visible, for 5 minutes straight; we recorded the emotional behaviors the child did as well as other behaviors and actions that the child did. After the 5 minutes, my partner and I took a minute to prepare for the time sampling observation. My partner and I had prepared codes before arriving to the classroom. The codes reflected the behaviors and reactions that are associated with emotional behaviors. The coding scheme was: 1-smiling, 2-neutral, 3-sad, 4-laughing, 5-angry, and 6-other. Once we were ready, we began recording emotional behaviors that the child did, in 30 second intervals, with 30-40 second intermission between the intervals. This was then repeated five times. The child that we observed was a quiet, reserved, four year old girl, who will be given the pseudo name Jay. Jay occasionally interacted with two other students that were 4-5 years of age; a quiet boy whose pseudo name will be Jack, and a social girl whose pseudo name will be Jess. When my partner and I arrived to the classroom, it was before lunch/snack time and the class was playing with different toys and activities around the room. A small handful, around five or six, girls were dressed up in princess costumes by a fairy tale doll house; one boy was dressed as a fireman and was walking around the room putting out “fires”. Two boys were on a mat playing with blocks, and Jay and Jack were playing with some of the toys that went along with the fire set.
Results
In the running record observation, my partner and I had a total number of six observations where we agreed on two of them, giving us a 33.3% agreement.
We agreed that Jay appeared to be neutral at the times 1:06 and 1:08. The observations that we disagreed on was when I observed Jay as being concerned at the times 1:06 and 1:07, whereas, my partner viewed Jay as just being neutral at time 1:07. Also, both my partner and I observed Jay smiling but we disagreed on the timeframe; I recorded her smiling at 1:08 whereas my partner recorded her smiling at 1:10. To an extent, my partner and I did observe the same behaviors, however, Jay would change her emotional reactions quickly, which made it hard to observe the features of the behavior. Also, I focused a little bit more on the behaviors that fell into the “other” category whereas my partner focused on the facial expressions which shows a difference in the emphases in our …show more content…
observations. In the time sampling observation, my partner and I had a total number of five observations and we completely agreed on each one, giving us a 100% agreement. The only behavior that we observed from Jay was that she was neutral during each interval, even if she was interacting with Jess, playing by herself, or watching others play around her. My partner and I observed the majority of same behaviors, my partner observed a few more actions than I did but she was also in a slightly better viewing position of Jay. Furthermore, we did not have any differences in emphasis during this observation. In research, one of the main goals of the data collected, is that the data is reliable and that the experiment can be redone and receive the same data.
In terms of reliability, inter-observer agreement is reliable when the percent of agreement is a high number because the number of agreements outweighed the number of disagreements. Thus, this means that there was a consistency of getting the same results. Out of the two naturalistic observations that my partner and I conducted, the one that led to the best inter-observer agreement was the time sampling method. This may be due to the fact that this method only focuses on one behavior and the coding system allowed us to be able to quickly record information and then redirect our attention back to Jay without potentially losing
information.
Discussion
The purpose of this study was to observe the emotional behavior of a child in a social and learning environment by using two naturalistic observation methods and determining which was more reliable. From this study, my partner and I observed that a child that is slow to warm up, tends to remain in a neutral emotion, since in both observations, Jay remained mainly neutral. Also, based off of the inter-observer agreement calculation, it can be concluded that the time sampling method is the most reliable observation method in this study. The relative benefits of using running record observations to record the behavior of a child includes the fact that the researcher can do a quantitative analysis on the frequency of a behavior that is produced by a child. For an example, Jay was neutral for approximately 3 minutes with one 2 second burst of a smile. Also, running record observations help the researcher visibly see a relation between different behaviors. For an example, when Jay was confused with a toy, her actions demonstrated that she was concerned by changing her facial features and becoming vocal about her concerns, this could potentially lead researcher to believe that there is a relationship to facial features and learning moments. The relative benefits of using time sampling observations to record the behavior of a child includes the fact that it makes it easier for the researcher to quickly record a behavior and then go back to observing the child. For an example, when Jay did a quick glimpse of a smile during the running record observation, it was difficult to acknowledge the change of the feature, then to quickly note it in the data. If this event were to have happened during the time sampling observation, my partner and I could have simply noted the code correlating to the behavior and then go back to watching Jay’s behavior.
This type of method observes the cause and effect of a behavior by making the behavior the dependent variable and having a control group, a group that doesn’t receive the change to produce the targeted behavior, and an experimental group, a group that does receive the change, and measure the effect of the change. The benefit of the naturalistic method is that, unlike the experimental method, the results show a more direct relation to a behavior and reality since experiments are conducted in an artificial standpoint, and more than one behavior can be observed. However, the experimental method allows researchers to study a specific behavior by isolating it, which also helps them to be able to control the situation and record more information, rather that losing it because the researcher was recording a non-targeted behavior. In this specific study that was conducted, the two naturalistic observation methods that were used included running record global observation and time sampling with predetermined codes. The running record method is when researchers records everything that happens including behaviors that are the main focus of the study. For an example, a researcher might be observing a child and wants to know their interactions with peers but they would also record interactions with teachers and other adults if it were to occur during the observation period. The time sampling method is when a researcher observes a desired behavior in set time intervals. The behavior is dissected into small actions related to the behavior, and are put into a coding system. For an example, if a researcher is observing how a child interactions with their peers, they would record actions related to this behavior, in small intervals-i.e. 30 seconds- and use codes to describe the behaviors, i.e. playing with peer=1, talking to peer=2, not interacting=3, etc. When looking for patterns in children’s overall social behaviors, using the running record method would be advantageous because it can help quantify the frequency time-wise, of the behaviors, thus helping to analyze how often a behavior occurs and how soon after another behavior occurs in response to the original behavior. The time sampling method would also be advantageous because it can help show the pattern of the behaviors interacting with each other in a sequence. The disadvantage of using the running record method is that the researcher can be overwhelmed by all of the behaviors that has to be captured and consequently misses information; for the time sampling method, the researcher can lose information showing the flow and actual relation between each behavior. If the researcher was only observing one specific behavior, the running record method would be advantageous because it could, again, show the time frequency of a behavior; likewise, the time sampling method could assist in quantifying the frequency of how often a behavior occurs in allotted intervals. However, the disadvantage with using the running record method is that the researcher would be losing extra information that can assist with learning more about a behavior. Also, with the time sampling method, the researcher would still lose information about the time of the occurrence of the behavior. The naturalistic observation that should give the highest level of inter-observer agreement should be the time sampling method. This assumption is due to the fact that this method has an agreed time frame; both of the partners are recording at the same time. Also, since we are looking at the same behavior with a code system, there shouldn’t be any discrepancies over the recorded behavior, and the coding system allows the researcher to be able to quickly take down more observations.
Method
The procedure for the running record observation method my partner and I used, was that we sat in an area where we were able to see the children but they were not able to see us in their field of vision. Then, once settled, we observed the child that was most visible, for 5 minutes straight; we recorded the emotional behaviors the child did as well as other behaviors and actions that the child did. After the 5 minutes, my partner and I took a minute to prepare for the time sampling observation. My partner and I had prepared codes before arriving to the classroom. The codes reflected the behaviors and reactions that are associated with emotional behaviors. The coding scheme was: 1-smiling, 2-neutral, 3-sad, 4-laughing, 5-angry, and 6-other. Once we were ready, we began recording emotional behaviors that the child did, in 30 second intervals, with 30-40 second intermission between the intervals. This was then repeated five times. The child that we observed was a quiet, reserved, four year old girl, who will be given the pseudo name Jay. Jay occasionally interacted with two other students that were 4-5 years of age; a quiet boy whose pseudo name will be Jack, and a social girl whose pseudo name will be Jess. When my partner and I arrived to the classroom, it was before lunch/snack time and the class was playing with different toys and activities around the room. A small handful, around five or six, girls were dressed up in princess costumes by a fairy tale doll house; one boy was dressed as a fireman and was walking around the room putting out “fires”. Two boys were on a mat playing with blocks, and Jay and Jack were playing with some of the toys that went along with the fire set.
Results
In the running record observation, my partner and I had a total number of six observations where we agreed on two of them, giving us a 33.3% agreement.
We agreed that Jay appeared to be neutral at the times 1:06 and 1:08. The observations that we disagreed on was when I observed Jay as being concerned at the times 1:06 and 1:07, whereas, my partner viewed Jay as just being neutral at time 1:07. Also, both my partner and I observed Jay smiling but we disagreed on the timeframe; I recorded her smiling at 1:08 whereas my partner recorded her smiling at 1:10. To an extent, my partner and I did observe the same behaviors, however, Jay would change her emotional reactions quickly, which made it hard to observe the features of the behavior. Also, I focused a little bit more on the behaviors that fell into the “other” category whereas my partner focused on the facial expressions which shows a difference in the emphases in our …show more content…
observations. In the time sampling observation, my partner and I had a total number of five observations and we completely agreed on each one, giving us a 100% agreement. The only behavior that we observed from Jay was that she was neutral during each interval, even if she was interacting with Jess, playing by herself, or watching others play around her. My partner and I observed the majority of same behaviors, my partner observed a few more actions than I did but she was also in a slightly better viewing position of Jay. Furthermore, we did not have any differences in emphasis during this observation. In research, one of the main goals of the data collected, is that the data is reliable and that the experiment can be redone and receive the same data.
In terms of reliability, inter-observer agreement is reliable when the percent of agreement is a high number because the number of agreements outweighed the number of disagreements. Thus, this means that there was a consistency of getting the same results. Out of the two naturalistic observations that my partner and I conducted, the one that led to the best inter-observer agreement was the time sampling method. This may be due to the fact that this method only focuses on one behavior and the coding system allowed us to be able to quickly record information and then redirect our attention back to Jay without potentially losing
information.
Discussion
The purpose of this study was to observe the emotional behavior of a child in a social and learning environment by using two naturalistic observation methods and determining which was more reliable. From this study, my partner and I observed that a child that is slow to warm up, tends to remain in a neutral emotion, since in both observations, Jay remained mainly neutral. Also, based off of the inter-observer agreement calculation, it can be concluded that the time sampling method is the most reliable observation method in this study. The relative benefits of using running record observations to record the behavior of a child includes the fact that the researcher can do a quantitative analysis on the frequency of a behavior that is produced by a child. For an example, Jay was neutral for approximately 3 minutes with one 2 second burst of a smile. Also, running record observations help the researcher visibly see a relation between different behaviors. For an example, when Jay was confused with a toy, her actions demonstrated that she was concerned by changing her facial features and becoming vocal about her concerns, this could potentially lead researcher to believe that there is a relationship to facial features and learning moments. The relative benefits of using time sampling observations to record the behavior of a child includes the fact that it makes it easier for the researcher to quickly record a behavior and then go back to observing the child. For an example, when Jay did a quick glimpse of a smile during the running record observation, it was difficult to acknowledge the change of the feature, then to quickly note it in the data. If this event were to have happened during the time sampling observation, my partner and I could have simply noted the code correlating to the behavior and then go back to watching Jay’s behavior.