A2 Psychology - Cambridge Mock Exam - Results and Feedback
Alan Nielsen
March 31, 2020
library(tidyverse)
library(ggthemes)
library(plyr)
library(kableExtra)
library(magrittr)
library(here)
library(ggrepel)
library(tippy)
theme_alan <- function(base_size = 12 , base_family = "")
{
half_line <- base_size/2
colors <- ggthemes_data$few
gray <- colors$medium["gray"]
black <- colors$dark["black"]
theme(
line = element_line(colour = "black", size = 0.5, linetype = 1, lineend = "butt"),
rect = element_rect(fill = "white",
colour = "black", size = 0.5, linetype = 1),
text = element_text(family = base_family, face = "plain", colour = "black",
size = base_size, lineheight = 0.9, hjust = 0.5, vjust = 0.5,
angle = 0, margin = margin(), debug = FALSE),
axis.line = element_blank(),
axis.line.x = NULL,
axis.line.y = NULL,
axis.text = element_text(size = rel(0.8), colour = "grey30"),
axis.text.x = element_text(margin = margin(t = 0.8 * half_line/2), vjust = 1),
axis.text.x.top = element_text(margin = margin(b = 0.8 * half_line/2), vjust = 0),
axis.text.y = element_text(margin = margin(r = 0.8 * half_line/2), hjust = 1),
axis.text.y.right = element_text(margin = margin(l = 0.8 * half_line/2), hjust = 0),
axis.ticks = element_line(colour = "grey20"),
axis.ticks.length = unit(half_line/2, "pt"),
axis.title.x = element_text(margin = margin(t = half_line), vjust = 1),
axis.title.x.top = element_text(margin = margin(b = half_line), vjust = 0),
axis.title.y = element_text(angle = 90, margin = margin(r = half_line), vjust = 1),
axis.title.y.right = element_text(angle = -90, margin = margin(l = half_line), vjust = 0),
legend.background = element_rect(colour = NA),
legend.spacing = unit(0.4, "cm"),
legend.spacing.x = NULL,
legend.spacing.y = NULL,
legend.margin = margin(0.2, 0.2, 0.2, 0.2, "cm"),
legend.key = element_rect(fill = "white", colour = NA),
legend.key.size = unit(1.2, "lines"),
legend.key.height = NULL,
legend.key.width = NULL,
legend.text = element_text(size = rel(0.8)),
legend.text.align = NULL,
legend.title = element_text(hjust = 0),
legend.title.align = NULL,
legend.position = "right",
legend.direction = NULL,
legend.justification = "center",
legend.box = NULL,
legend.box.margin = margin(0, 0, 0, 0, "cm"),
legend.box.background = element_blank(),
legend.box.spacing = unit(0.4, "cm"),
panel.background = element_rect(fill = "white", colour = NA),
panel.border = element_rect(fill = NA, colour = "grey20"),
panel.grid.major = element_line(colour = "grey92"),
panel.grid.minor = element_line(colour = "grey92", size = 0.25),
panel.spacing = unit(half_line, "pt"),
panel.spacing.x = NULL,
panel.spacing.y = NULL,
panel.ontop = FALSE,
strip.background = element_rect(fill = "NA", colour = "NA"),
strip.text = element_text(colour = "grey10", size = rel(0.8)),
strip.text.x = element_text(margin = margin(t = half_line, b = half_line)),
strip.text.y = element_text(angle = 0, margin = margin(l = half_line, r = half_line)),
strip.placement = "inside",
strip.placement.x = NULL,
strip.placement.y = NULL,
strip.switch.pad.grid = unit(0.1, "cm"),
strip.switch.pad.wrap = unit(0.1, "cm"),
plot.background = element_rect(colour = "white"),
plot.title = element_text(size = rel(1.2), hjust = 0, vjust = 1, margin = margin(b = half_line * 1.2)),
plot.subtitle = element_text(size = rel(0.9), hjust = 0, vjust = 1, margin = margin(b = half_line * 0.9)),
plot.caption = element_text(size = rel(0.9), hjust = 1, vjust = 1, margin = margin(t = half_line * 0.9)),
plot.margin = margin(half_line, half_line, half_line, half_line),
complete = TRUE)
}
# Wrapper Function for Long Graph Titles
wrapper <- function(x, ...)
{
paste(strwrap(x, ...), collapse = "\n")
}
pd <- position_dodge(width = 0.8) #My standard dodging for graphs
#RDFZ Reds
RDFZPink <- "#cf8f8d"
RDFZRed1 <- "#ae002b"
RDFZRed2 <- "#991815"
RDFZRed3 <- "#78011e"
RDFZRed4 <- "#4b0315"
#Grade Colors
GradeColors <- c("#8900df", "#0092df", "#00df76", "#94df00", "#94df00", "#df5300", "#9b0f00")
#function for allowing inline code chunks to be shown verbatim
rinline <- function(code){
html <- '<code class="r">``` `r CODE` ```</code>'
sub("CODE", code, html)
}## Read in the Data File
MockGrades <- read.csv(here("MockGrades.csv"))
## Make a Unique Column for Grouping
MockGrades %<>%
unite(Combo, Paper, Teacher, Student, Anon, sep = "-", remove = FALSE)
## Split by Paper (this obtains us some values we need)
Paper3 <- subset(MockGrades, Paper == 3)
Paper4 <- subset(MockGrades, Paper == 4)
## Obtain Maximum Values (total score of test)
P3MaxVal <- as.numeric(sum(head(Paper3, length(unique(Paper3$Question)))$Value))
P4MaxVal <- as.numeric(sum(head(Paper4, length(unique(Paper4$Question)))$Value))
## Summarize the Data for Each Participant
Paper3Totals <-
Paper3 %>%
group_by(Combo) %>%
dplyr::summarise(sum = sum(Grade)) %>%
mutate(Grade = round(sum/P3MaxVal*100, 1)) %>%
separate(Combo, into = c("Paper", "Teacher", "Student", "Anon"), sep = "-", remove = TRUE) %>%
arrange(Student)
Paper4Totals <-
Paper4 %>%
group_by(Combo) %>%
dplyr::summarise(sum = sum(Grade)) %>%
mutate(Grade = round(sum/P4MaxVal*100, 1)) %>%
separate(Combo, into = c("Paper", "Teacher", "Student", "Anon"), sep = "-", remove = TRUE) %>%
arrange(Student) #FILL THIS IN WHEN SETTING UP THE SHEET - IT WILL HELP WITH NAMING THE TABLES AND IMAGES THAT ARE OUTPUT, SAVING YOU TIME
Class <- "A2 Psychology" #Name of the Class Goes Here
Eval <- "Cambridge Mock Exam" #Name of what is being evaluated goes herePaper 3
Questions
Paper 3 consisted of questions testing student’s knowledge of A2 Psychology. The questions and their values can be seen in the table below.
##KNITTED TABLE OF QUESTIONS AND POINT VALUES
QTitle <- paste(Class, Eval, "Paper 3", "Questions and Values", sep = " - ")
QT3 <-
Paper3 %>%
head(length(unique(Paper3$Question))) %>%
subset(select = c(Qnum, Question, Value)) %>%
setNames(c("Number", "Question", "Value"))
QuestionsTableP3 <-
QT3 %>%
knitr::kable(caption = QTitle, row.names = F) %>%
row_spec(0, bold = T, color = "white", background = RDFZRed3)%>%
kable_styling(full_width = FALSE,
bootstrap_options = c("striped", "hover", "condensed"),
fixed_thead = TRUE)
save_kable(QuestionsTableP3, paste(QTitle, ".png", sep = ""))
QuestionsTableP3| Number | Question | Value |
|---|---|---|
| 1 | Explain what is meant by the term ‘personal space’ | 2 |
| 2 | Describe McCarthy’s ‘4Ps marketing mix’ model of advertising | 4 |
| 3 | Explain two weaknesses of system 1 (fast type) / system 2 (slow-type) thinking as categorised by Schleifer, 2012 | 6 |
| 4 | Describe what psychologists have discovered about choice heuristics in consumer decision-making (availability/representativeness, anchoring and purchase quantity decisions, pre-cognitive decisions) | 8 |
| 5 | Explain what is meant by “learned helplessness” | 2 |
| 6 | Describe the token economy as a treatment for schizophrenia and delusional disorder | 4 |
| 7 | Explain one similarity and one difference between drug treatments for depression and cognitive restructuring treatment for depression | 6 |
| 8 | Evaluate explanations of anxiety, including a discussion of nature versus nurture | 10 |
Overall Performance
#Single Histogram
OverallTitle <- paste(Class, Eval, "Paper 3", "Histogram of Grades", sep = " - ")
ggplot(data=Paper3Totals, aes(x=Grade)) +
geom_histogram(aes(y=..density..), alpha = 1, position = "identity", fill = RDFZRed2) +
labs(x="Grade (Percentage)", y="Density") +
stat_function(fun=dnorm, args = list(mean=mean(Paper3Totals$Grade), sd=sd(Paper3Totals$Grade)),
color=RDFZRed4, size = 1.4) +
scale_x_continuous(limits = c(0,100)) +
theme_alan() +
theme(legend.position = "none") +
ggtitle(wrapper(OverallTitle, width = 45))
ggsave(here(paste(OverallTitle, ".png", sep = "")), plot = last_plot(), device = NULL, path = NULL,
width = 10, height = 6, units = c("in", "cm", "mm"),
dpi = 600)Overall performance can be seen in the graph above. We had a fairly normal distribution of grades (Shapiro-Wilk Normality test: W= 0.93, p= 0.056) with a mean score of 76.9% (Median = 81%) and a standard deviation of 16.4%.
The high score on the test was 41 out of 42 marks (97.6%), which was achieved by 3 student(s).
The low score on the test was 17 out of 42 marks (40.5%), which was achieved by 1 student(s).
Letter Grades (Raw)
Typical raw letter grade boundaries for the school can be seen in the table below:
#Note that this already includes rounding up any grade above 0.5 to the next letter grade (e.g. 79.5% = A)
Letters <- c("A*", "A", "B", "C", "D", "E", "U")
MinVal <- c(89.5, 79.5, 69.5, 59.5, 49.5, 39.5, 0)
MaxVal <- c (100, 89.49, 79.49, 69.49, 59.49, 49.49, 39.49)
LetterBoundaries <-
cbind.data.frame(Letters, MinVal, MaxVal) %>%
setNames(c("Letter", "Bottom", "Top"))
LetterGradesTable <-
LetterBoundaries %>%
setNames(c("Letter Grade", "Bottom Boundary", "Top Boundary")) %>%
knitr::kable(caption = "RDFZ Letter Grade Boundaries", row.names = F) %>%
row_spec(0, bold = T, color = "white", background = RDFZRed3)%>%
kable_styling(full_width = FALSE,
bootstrap_options = c("striped", "hover", "condensed"),
fixed_thead = TRUE) %>%
footnote(general = "Note this includes rounding of grades like 79.5% up a grade boundary" )
LetterGradesTable| Letter Grade | Bottom Boundary | Top Boundary |
|---|---|---|
| A* | 89.5 | 100.00 |
| A | 79.5 | 89.49 |
| B | 69.5 | 79.49 |
| C | 59.5 | 69.49 |
| D | 49.5 | 59.49 |
| E | 39.5 | 49.49 |
| U | 0.0 | 39.49 |
| Note: | ||
| Note this includes rounding of grades like 79.5% up a grade boundary |
If we assign grade boundaries based on these we can see the following distribution of grades
GB1Title <- paste(Class, Eval, "Paper 3", "Letter Grades (Raw)", sep = " - ")
Paper3Totals %<>%
mutate(Letter = cut(x= Paper3Totals$Grade,
breaks = c(LetterBoundaries$Bottom,100),
labels= map_df(LetterBoundaries,rev)$Letter)) %>%
mutate(Letter = factor(Letter, levels =c("U", "E", "D", "C", "B", "A", "A*")))
ggplot(data=Paper3Totals, aes(x=Letter, fill = Letter)) +
geom_bar(stat = "count", position = pd, width = 0.8) +
scale_x_discrete(drop = FALSE) +
scale_fill_manual(values = (c(rev(GradeColors[1:6])) )) + #Janky because of missing grade boundaries (no students in bins)
labs(x="Letter Grade", y="Count") +
theme_alan() +
ggtitle(wrapper(GB1Title, width = 45))
ggsave(here(paste(GB1Title, ".png", sep = "")), plot = last_plot(), device = NULL, path = NULL,
width = 8, height = 6, units = c("in", "cm", "mm"),
dpi = 600)
LetterCounts <- count(Paper3Totals$Letter)
LetterBoundaries2 <-
LetterBoundaries %>%
mutate(Letter = factor(Letter, levels =c("U", "E", "D", "C", "B", "A", "A*"))) %>%
mutate(Count = plyr::mapvalues(Letter, from = LetterCounts$x, to = LetterCounts$freq)) %>%
mutate(Count = as.numeric(as.character(Count))) %>%
mutate(Count = replace_na(Count, 0))This is actually a pretty good distribution of Grade Boundaries, suggesting that performance on this assessment was at a pretty high level: A total of 15 students earned an A or A-star, with only 2 students earning an E or a U on the exam. The total counts of students in each grade category can be seen in the table below
LetterBoundaries2 %>%
setNames(c("Letter Grade", "Bottom Boundary", "Top Boundary", "Count")) %>%
knitr::kable(caption = "RDFZ Letter Grade Boundaries", row.names = F) %>%
row_spec(0, bold = T, color = "white", background = RDFZRed3)%>%
kable_styling(full_width = FALSE,
bootstrap_options = c("striped", "hover", "condensed"),
fixed_thead = TRUE) | Letter Grade | Bottom Boundary | Top Boundary | Count |
|---|---|---|---|
| A* | 89.5 | 100.00 | 7 |
| A | 79.5 | 89.49 | 8 |
| B | 69.5 | 79.49 | 5 |
| C | 59.5 | 69.49 | 3 |
| D | 49.5 | 59.49 | 4 |
| E | 39.5 | 49.49 | 2 |
| U | 0.0 | 39.49 | 0 |
Between Class Performance
How do the two classes for A2 Psychology compare to one another on Cambridge Mock Exam - Paper 3?
Below, you can see a histogram of performance between the two classes.
BetweenTitle <- paste(Class, Eval, "Paper 3", "Histogram of Grades by Class", sep = " - ")
#Histograms by Class
JuneP3Grades <- subset(Paper3Totals, Teacher == "June")
AlanP3Grades <- subset(Paper3Totals, Teacher == "Alan")
ggplot(data=Paper3Totals, aes(x=Grade, fill = Teacher)) +
geom_histogram(aes(y=..density..), alpha = 0.75, position = "identity") +
labs(x="Grade (Percentage)", y="Density") +
stat_function(fun=dnorm, args = list(mean=mean(JuneP3Grades$Grade), sd=sd(JuneP3Grades$Grade)), color=RDFZRed1, size = 1.2) +
stat_function(fun=dnorm, args = list(mean=mean(AlanP3Grades$Grade), sd=sd(AlanP3Grades$Grade)), color=RDFZRed4, size = 1.2) +
scale_fill_manual(values= c(RDFZRed4, RDFZRed1)) +
scale_x_continuous(limits = c(0,100)) +
theme_alan() +
ggtitle(wrapper(BetweenTitle, width = 45))
ggsave(here(paste(BetweenTitle, ".png", sep = "")), plot = last_plot(), device = NULL, path = NULL,
width = 10, height = 4, units = c("in", "cm", "mm"),
dpi = 600)
testcompP3 <- t.test(JuneP3Grades$Grade, AlanP3Grades$Grade, warning = FALSE, message = FALSE)We can see that there is a difference of 10% between the classes. Alan’s A2 Psychology class averaged a grade of 83.1% (sd = 8.9%), while June’s class averaged a grade of 73.1% (sd = 18.9%). This difference is not significant: t(25.8) = -1.92, p= 0.07.
Paper 4
Questions
Paper 4 consisted of 8 Questions testing student’s knowledge of the application of research methodology. The questions and their values can be seen in the table below.
##KNITTED TABLE OF QUESTIONS AND POINT VALUES
QTitle2 <- paste(Class, Eval, "Paper 3", "Questions and Values", sep = " - ")
QT4 <-
Paper4 %>%
head(length(unique(Paper4$Question))) %>%
subset(select = c(Qnum, Question, Value)) %>%
setNames(c("Number", "Question", "Value"))
QuestionsTableP4 <-
QT4 %>%
knitr::kable(caption = QTitle, row.names = F) %>%
row_spec(0, bold = T, color = "white", background = RDFZRed3)%>%
kable_styling(full_width = FALSE,
bootstrap_options = c("striped", "hover", "condensed"),
fixed_thead = TRUE)
save_kable(QuestionsTableP4, paste(QTitle, ".png", sep = ""))
QuestionsTableP4| Number | Question | Value |
|---|---|---|
| 1 | Explain what is meant by “generalised anxiety disorder” | 2 |
| 2 | Suggest one advantage and one disadvantage of using quantitative data to assess anxiety | 4 |
| 3 | Ina study by Milgram et al., a stooge (confederate) intruded into a line of people waiting in a queue. “Hey buddy, we’ve been waiting. Get off the line and go to the back!” was one of the responses to the intrusion. Outline the research method used in this study | 2 |
| 4 | The behavior of people in the queue during the intrusion was observed in three response categories. Identify two response categories used, other than verbal objections | 2 |
| 5 | Discuss the strengths and weaknesses of the cognitive explanations of schizophrenia. You should include a conclusion in your answer. | 5 |
| 6 | The feeling-state theory proposes that impulse control disorders are caused by intense positive feelings that become associated with an experience such as stealing or gambling. Design an experiment to investigate the intensity of positive feelings about an event, inpeople with and without an impulse control disorder. | 10 |
| 7 | Explain the psychological and methodological evidence on which your experiment is based | 8 |
| 8 | “Field experiments on consumer behaviour conducted in one country, such as that on choice blindness by Hall et al. (2010) cannot be generalised to other countries.”. To what extent do you agree with this statement? Use examples of research you have studied to support your answer. | 12 |
Overall Performance
#Single Histogram
OverallTitle2 <- paste(Class, Eval, "Paper 4", "Histogram of Grades", sep = " - ")
ggplot(data=Paper4Totals, aes(x=Grade)) +
geom_histogram(aes(y=..density..), alpha = 1, position = "identity", fill = RDFZRed2) +
labs(x="Grade (Percentage)", y="Density") +
stat_function(fun=dnorm, args = list(mean=mean(Paper4Totals$Grade), sd=sd(Paper4Totals$Grade)),
color=RDFZRed4, size = 1.4) +
scale_x_continuous(limits = c(0,100)) +
theme_alan() +
theme(legend.position = "none") +
ggtitle(wrapper(OverallTitle2, width = 45))
ggsave(here(paste(OverallTitle2, ".png", sep = "")), plot = last_plot(), device = NULL, path = NULL,
width = 10, height = 6, units = c("in", "cm", "mm"),
dpi = 600)Overall performance can be seen in the graph above. We had a fairly normal distribution of grades (Shapiro-Wilk Normality test: W= 0.95, p= 0.204) with a mean score of 70.3% (Median = 73.3%) and a standard deviation of 15.1%.
The high score on the test was 43 out of 45 marks (95.6%), which was achieved by 1 student(s).
The low score on the test was 20 out of 45 marks (44.4%), which was achieved by 3 student(s).
Letter Grades (Raw)
Typical raw letter grade boundaries for the school can be seen in the table below:
LetterGradesTable| Letter Grade | Bottom Boundary | Top Boundary |
|---|---|---|
| A* | 89.5 | 100.00 |
| A | 79.5 | 89.49 |
| B | 69.5 | 79.49 |
| C | 59.5 | 69.49 |
| D | 49.5 | 59.49 |
| E | 39.5 | 49.49 |
| U | 0.0 | 39.49 |
| Note: | ||
| Note this includes rounding of grades like 79.5% up a grade boundary |
If we assign grade boundaries based on these we can see the following distribution of grades
GB1Title2 <- paste(Class, Eval, "Paper 3", "Letter Grades (Raw)", sep = " - ")
Paper4Totals %<>%
mutate(Letter = cut(x= Paper4Totals$Grade,
breaks = c(LetterBoundaries$Bottom,100),
labels= map_df(LetterBoundaries,rev)$Letter)) %>%
mutate(Letter = factor(Letter, levels =c("U", "E", "D", "C", "B", "A", "A*")))
ggplot(data=Paper4Totals, aes(x=Letter, fill = Letter)) +
geom_bar(stat = "count", position = pd, width = 0.8) +
scale_x_discrete(drop = FALSE) +
scale_fill_manual(values = (c(GradeColors[7], rev(GradeColors[1:5])) )) + #Janky because of missing grade boundaries (no students in bins)
labs(x="Letter Grade", y="Count") +
theme_alan() +
ggtitle(wrapper(GB1Title2, width = 45))
ggsave(here(paste(GB1Title2, ".png", sep = "")), plot = last_plot(), device = NULL, path = NULL,
width = 8, height = 6, units = c("in", "cm", "mm"),
dpi = 600)
LetterCounts2 <- count(Paper4Totals$Letter)
LetterBoundaries2 %<>%
mutate(Letter = factor(Letter, levels =c("U", "E", "D", "C", "B", "A", "A*"))) %>%
mutate(Count2 = plyr::mapvalues(Letter, from = LetterCounts2$x, to = LetterCounts2$freq)) %>%
mutate(Count2 = as.numeric(as.character(Count2))) %>%
mutate(Count2 = replace_na(Count2, 0))Performance on Paper 4 was pretty disappointing: A total of of 9 students earned an A or A-star, but 4 students earned an E or a U on the exam. The total counts of students in each grade category can be seen in the table below
LetterBoundaries2 %>%
setNames(c("Letter Grade", "Bottom Boundary", "Top Boundary", "Paper 3", "Paper 4")) %>%
knitr::kable(caption = "RDFZ Letter Grade Boundaries", row.names = F) %>%
row_spec(0, bold = T, color = "white", background = RDFZRed3)%>%
kable_styling(full_width = FALSE,
bootstrap_options = c("striped", "hover", "condensed"),
fixed_thead = TRUE) | Letter Grade | Bottom Boundary | Top Boundary | Paper 3 | Paper 4 |
|---|---|---|---|---|
| A* | 89.5 | 100.00 | 7 | 2 |
| A | 79.5 | 89.49 | 8 | 7 |
| B | 69.5 | 79.49 | 5 | 7 |
| C | 59.5 | 69.49 | 3 | 6 |
| D | 49.5 | 59.49 | 4 | 3 |
| E | 39.5 | 49.49 | 2 | 4 |
| U | 0.0 | 39.49 | 0 | 0 |
Between Class Performance
How do the two classes for A2 Psychology compare to one another on Cambridge Mock Exam - Paper 4?
Below, you can see a histogram of performance between the two classes.
BetweenTitle2 <- paste(Class, Eval, "Paper 4", "Histogram of Grades by Class)", sep = " - ")
#Histograms by Class
JuneP4Grades <- subset(Paper4Totals, Teacher == "June")
AlanP4Grades <- subset(Paper4Totals, Teacher == "Alan")
ggplot(data=Paper4Totals, aes(x=Grade, fill = Teacher)) +
geom_histogram(aes(y=..density..), alpha = 0.75, position = "identity") +
labs(x="Grade (Percentage)", y="Density") +
stat_function(fun=dnorm, args = list(mean=mean(JuneP4Grades$Grade), sd=sd(JuneP4Grades$Grade)), color=RDFZRed1, size = 1.2) +
stat_function(fun=dnorm, args = list(mean=mean(AlanP4Grades$Grade), sd=sd(AlanP4Grades$Grade)), color=RDFZRed4, size = 1.2) +
scale_fill_manual(values= c(RDFZRed4, RDFZRed1)) +
scale_x_continuous(limits = c(0,100)) +
theme_alan() +
ggtitle(wrapper(BetweenTitle2, width = 45))
ggsave(here(paste(BetweenTitle2, ".png", sep = "")), plot = last_plot(), device = NULL, path = NULL,
width = 10, height = 4, units = c("in", "cm", "mm"),
dpi = 600)
testcompP4 <- t.test(JuneP4Grades$Grade, AlanP4Grades$Grade, warning = FALSE, message = FALSE)We can see that there isn’t much difference at all between the classes. Alan’s A2 Psychology class averaged a grade of 72.2% (sd = 13%), while June’s class averaged a grade of 69.4% (sd = 16.3%). This difference is not significant: t(22.4) = -0.52, p= 0.61.
Between Paper Comparison
It should be fairly obvious from the above analyses that the results of Paper 3 and Paper 4 were quite different from one another. Recall that the average performance on Paper 3 was 81%, while for Paper 4 the average was 73%. The distributions of grades in the two papers can be seen relative to one another in the Histogram below:
#Creating a Dataframe with Scores for both tests
MockTotals <- rbind.data.frame(Paper3Totals, Paper4Totals)
ComparisonTitle <- paste(Class, Eval, "Comparison of Paper 3 and Paper 4", sep = " - ")
#Histograms
ggplot(data=MockTotals, aes(x=Grade, fill = Paper)) +
geom_histogram(aes(y=..density..), alpha = 0.6, position = "identity") +
labs(x="Grade (Percentage)", y="Density") +
stat_function(fun=dnorm, args = list(mean=mean(Paper3Totals$Grade), sd=sd(Paper3Totals$Grade)), color=RDFZRed4, size = 1.2) +
stat_function(fun=dnorm, args = list(mean=mean(Paper4Totals$Grade), sd=sd(Paper4Totals$Grade)), color=RDFZRed1, size = 1.2) +
scale_fill_manual(values= c(RDFZRed4, RDFZRed1)) +
scale_x_continuous(limits = c(0,100)) +
theme_alan() +
ggtitle(wrapper(ComparisonTitle, width = 45))
ggsave(here(paste(ComparisonTitle, ".png", sep = "")), plot = last_plot(), device = NULL, path = NULL,
width = 10, height = 4, units = c("in", "cm", "mm"),
dpi = 600)
#Adding Scores for Omnibus Score
TestMaxVal <- P3MaxVal + P4MaxVal
MockTotals2 <-
Paper3Totals %>%
mutate(sum2 = as.numeric(plyr::mapvalues(Student, from = Paper4Totals$Student, to = Paper4Totals$sum))) %>%
mutate(Grade2 = as.numeric(plyr::mapvalues(Student, from = Paper4Totals$Student, to = Paper4Totals$Grade))) %>%
setNames(c("Paper", "Teacher", "Student", "Anon", "Score1", "Grade1", "Letter1", "Score2", "Grade2")) %>%
mutate(Score3 = Score1 + Score2) %>%
mutate(Grade3 = round(Score3/TestMaxVal*100, 1)) %>%
mutate(Letter = cut(x= Grade3,
breaks = c(LetterBoundaries$Bottom,100),
labels= map_df(LetterBoundaries,rev)$Letter)) %>%
mutate(Letter = factor(Letter, levels =c("U", "E", "D", "C", "B", "A", "A*"))) %>%
mutate(Diff = Grade1 - Grade2)
#T test comparing values
testcompP3P4 <- t.test(MockTotals2$Grade1, MockTotals2$Grade2, paired = TRUE, warning = FALSE, message = FALSE)
#Correlation of Values
P3P4Corr <- cor.test(MockTotals2$Grade1, MockTotals2$Grade2)The difference between the scores on the two tests is significant: t(28) = 3.8, p= 0.001. This is a highly robust difference - of the 29 students in A2 Psychology, a total of 22 scored lower on Paper 4 than on Paper 3 by an average of 10.1 percent, compared to the 6 students who scored higher on Paper 4 by an average of 5.2 percent.
Despite Paper 4 being more difficult than Paper 3, scores between the two were highly correlated with one another (Pearson’s r= 0.83; t(27)= 7.64, p= <0.001). A graph of the correlation can be seen below:
ComparisonTitle2 <- paste(Class, Eval, "Correlation of Paper 3 and Paper 4 Scores", sep = " - ")
ggplot(data=MockTotals2, aes(x=Grade1, y = Grade2)) +
geom_point(color = RDFZRed3, size = 2) +
geom_smooth(method = "lm", se = FALSE, size = 1.2, color = RDFZRed4) +
geom_label_repel(aes(label = Anon)) +
scale_x_continuous(breaks = c(25, 50, 75, 100), limits= c(20,100) ) +
scale_y_continuous(breaks = c(25, 50, 75, 100), limits= c(20,100) ) +
labs(x="Paper 3 Grade", y="Paper 4 Grade") +
theme_alan() +
ggtitle(wrapper(ComparisonTitle2, width = 45))
ggsave(here(paste(ComparisonTitle2, ".png", sep = "")), plot = last_plot(), device = NULL, path = NULL,
width = 6, height = 6, units = c("in", "cm", "mm"),
dpi = 600)The difficulty of Paper 4 relative to Paper 3 is a bit surprising - for A2 Psychology results are typically pretty similar between the two papers, with Paper 4 actually being slightly easier. This might suggest that we chose a difficult set of questions for Paper 4, or that we have given our students less practice handling Paper 4 style questions. Likely both of these are true.
Grade Correction
There are broadly two ways that we can correct grades on these tests. First, we can correct for any mistakes we made as teachers - either by giving unfair questions, questions that are too hard, or by not preparing our students adequately for certain types of questions. The second is to curve the entire grade upward to be in line with achievement goals for the course and to align our results with what students could expect from their actual CAIE exams.
Curving Paper 4
As noted above, our students performed significantly worse on Paper 4 than on Paper 3, counter to what is typical with CAIE examinations in A2 Psychology. As such, we will be “correcting” Paper 4 Grades to make them more in line with the grades from Paper 3. On average, students scored 6.6 percent higher (out of 42) on Paper 3. So, we awarded all students an additional 3 marks on their Paper 4 scores (capped at a maximum score of 45).
This change results in the Distribution of Grades seen below:
#Curving and Assigning Letter Boundaries
MockTotalsCurved <-
MockTotals2 %>%
subset(select = c("Teacher", "Student", "Anon", "Score1", "Score2")) %>%
mutate(Score2 = Score2 +3) %>%
mutate(Score2 = ifelse(Score2 >P4MaxVal, P4MaxVal, Score2)) %>%
mutate(SumScore = Score1 + Score2) %>%
mutate(Grade = round(SumScore/TestMaxVal*100,1)) %>%
mutate(Letter = cut(x= Grade,
breaks = c(LetterBoundaries$Bottom,100),
labels= map_df(LetterBoundaries,rev)$Letter)) %>%
mutate(Letter = factor(Letter, levels =c("U", "E", "D", "C", "B", "A", "A*")))
#Plotting Distribution of Letter Grades
CurveTitle1 <- paste(Class, Eval, "Distribution of Grades after Curving Paper 4", sep = " - ")
ggplot(data=MockTotalsCurved, aes(x=Letter, fill = Letter)) +
geom_bar(stat = "count", position = pd, width = 0.8) +
scale_x_discrete(drop = FALSE) +
scale_fill_manual(values = (c(rev(GradeColors[1:6])) )) + #Janky because of missing grade boundaries (no students in bins)
labs(x="Letter Grade", y="Count") +
theme_alan() +
ggtitle(wrapper(CurveTitle1, width = 45))
ggsave(here(paste(CurveTitle1, ".png", sep = "")), plot = last_plot(), device = NULL, path = NULL,
width = 8, height = 6, units = c("in", "cm", "mm"),
dpi = 600)Unsurprisingly, when we curve this up the overall distribution looks a lot like the distribution did for Paper 3 alone. So there isn’t much to say about this.
Applying an Overall Curve
Rather than correcting for poor performance on individual questions, Cambridge curved performance on both Papers as a whole (for letter Grades not include A-star) and for overall performance.
Here, we will take the second of these approaches.
It should be noted that for SY Option students (A2 Psychology Students taking only Paper 3 and Paper 4) Cambridge does not set an A-star level. We will, however, do so here.
In 2019, Cambridge set the boundaries below for their grade levels:
Letters <- c("A*", "A", "B", "C", "D", "E", "U")
MinVal <- c(94, 85, 75, 66, 57, 49, 0)
MaxVal <- c(120, 93, 84, 74, 65, 56, 48)
CambridgeBoundaries <-
cbind.data.frame(Letters, round(MinVal/120*100,2), round(MaxVal/120*100,2)) %>%
setNames(c("Letter", "Bottom", "Top"))
TableNote <- "These are transformed from the Raw Score Boundaries out of 120 marks from the 2019 Cambridge Psychology Exam" #to be wrapped
CambridgeGradesTable <-
CambridgeBoundaries %>%
setNames(c("Letter Grade", "Bottom Boundary", "Top Boundary")) %>%
knitr::kable(caption = "Cambridge Letter Grade Boundaries", row.names = F) %>%
row_spec(0, bold = T, color = "white", background = RDFZRed3)%>%
kable_styling(full_width = FALSE,
bootstrap_options = c("striped", "hover", "condensed"),
fixed_thead = TRUE) %>%
footnote(general = wrapper(TableNote, width = 40) )
CambridgeGradesTable| Letter Grade | Bottom Boundary | Top Boundary |
|---|---|---|
| A* | 78.33 | 100.00 |
| A | 70.83 | 77.50 |
| B | 62.50 | 70.00 |
| C | 55.00 | 61.67 |
| D | 47.50 | 54.17 |
| E | 40.83 | 46.67 |
| U | 0.00 | 40.00 |
| Note: | ||
|
These are transformed from the Raw Score Boundaries out of 120 marks from the 2019 Cambridge Psychology Exam |
These move every grade boundary down by approximately 10 percent. If we were to apply this curve to our A2 Psychology grades, they would be distributed thusly:
#Curving and Assigning Letter Boundaries
MockTotalsCurved %<>%
mutate(`Letter (Revised- Cambridge)` = cut(x= Grade,
breaks = c(CambridgeBoundaries$Bottom,100),
labels= map_df(CambridgeBoundaries,rev)$Letter)) %>%
mutate(`Letter (Revised- Cambridge)` = factor(`Letter (Revised- Cambridge)`, levels =c("U", "E", "D", "C", "B", "A", "A*")))
#Plotting Distribution of Letter Grades
CurveTitle2 <- paste(Class, Eval, "Distribution of Grades after Applying Full Cambridge Curve", sep = " - ")
ggplot(data=MockTotalsCurved, aes(x=`Letter (Revised- Cambridge)`, fill = `Letter (Revised- Cambridge)`)) +
geom_bar(stat = "count", position = pd, width = 0.8) +
scale_x_discrete(drop = FALSE) +
scale_fill_manual(values = (c(rev(GradeColors[1:5])) )) + #Janky because of missing grade boundaries (no students in bins)
labs(x="Letter Grade (Cambridge Curved)", y="Count") +
theme_alan() +
ggtitle(wrapper(CurveTitle2, width = 45))
ggsave(here(paste(CurveTitle2, ".png", sep = "")), plot = last_plot(), device = NULL, path = NULL,
width = 8, height = 6, units = c("in", "cm", "mm"),
dpi = 600)This would almost certainly be too much of a curve for a variety of reasons
- We almost certainly grade the longer-format questions easier than Cambridge does (i.e. we doubt that Cambridge gives many 10s to Evaluate/Design questions, but obviously with Cambridge’s data never being opened we cannot know)
- We have already curved up the Paper 4 Grades
- Students were provided with a number of practice tests, included in which were some of the exact questions given on these Mock Exams
- We want students to remain motivated to continue to study for their CAIE examination
Collectively these three things gave students an advantage over their probable Cambridge Exam grades. Because of this, we chose to apply a lesser curve (and one that would help our lowest-performing student, who was unaffected by the Cambridge Curve) and to award all of our students an additional 5 marks, capped at a total of 87 (100%). After applying these corrections we kept the actual grade boundaries at their standard school values (e.g. A* = 90%+).
The distribution of grades after this adjustment can be seen below:
#Curving and Assigning Letter Boundaries
MockTotalsCurvedFinal <-
MockTotalsCurved %>%
mutate(SumScore = SumScore + 5) %>%
mutate(SumScore = ifelse(SumScore > TestMaxVal, TestMaxVal, SumScore)) %>%
mutate(Grade = round(SumScore/TestMaxVal*100,1)) %>%
mutate(`Letter (Revised)` = cut(x= Grade,
breaks = c(LetterBoundaries$Bottom,100),
labels= map_df(LetterBoundaries,rev)$Letter)) %>%
#mutate(`Letter (Revised)` = factor(`Letter (Revised)`, levels =c("U", "E", "D", "C", "B", "A", "A*"))) %>%
subset(select = c("Teacher", "Student", "Anon", "Score1", "Score2", "SumScore", "Grade", "Letter (Revised)")) %>%
setNames(c("Teacher", "Student", "Anon", "Paper 3", "Paper 4 (R)", "Total (R)", "Grade (R)", "Letter"))
#Plotting Distribution of Letter Grades
CurveTitle3 <- paste(Class, Eval, "Distribution of Grades after Final Curve", sep = " - ")
ggplot(data=MockTotalsCurvedFinal, aes(x=`Letter`, fill = `Letter`)) +
geom_bar(stat = "count", position = pd, width = 0.8) +
scale_x_discrete(drop = FALSE) +
scale_fill_manual(values = (c(rev(GradeColors[1:5])) )) + #Janky because of missing grade boundaries (no students in bins)
labs(x="Letter Grade (Revised)", y="Count") +
theme_alan() +
ggtitle(wrapper(CurveTitle3, width = 45))
ggsave(here(paste(CurveTitle3, ".png", sep = "")), plot = last_plot(), device = NULL, path = NULL,
width = 8, height = 6, units = c("in", "cm", "mm"),
dpi = 600)Tables
Tables for Engage
# Alan
AlanTitle <- paste(Class, Eval, "Final Grades - Alan", sep = " - ")
AlanTable <-
MockTotalsCurvedFinal %>%
subset(Teacher == "Alan") %>%
subset(select = -Teacher) %>%
arrange(-`Grade (R)`) %>%
knitr::kable(caption = AlanTitle, row.names = F) %>%
row_spec(0, bold = T, color = "white", background = RDFZRed3)%>%
kable_styling(full_width = FALSE,
bootstrap_options = c("striped", "hover", "condensed"),
fixed_thead = TRUE)
save_kable(AlanTable, paste(AlanTitle, ".png", sep = ""))
# June
JuneTitle <- paste(Class, Eval, "Final Grades - June", sep = " - ")
JuneTable <-
MockTotalsCurvedFinal %>%
subset(Teacher == "June") %>%
subset(select = -Teacher) %>%
arrange(-`Grade (R)`) %>%
knitr::kable(caption = JuneTitle, row.names = F) %>%
row_spec(0, bold = T, color = "white", background = RDFZRed3)%>%
kable_styling(full_width = FALSE,
bootstrap_options = c("striped", "hover", "condensed"),
fixed_thead = TRUE)
save_kable(JuneTable, paste(JuneTitle, ".png", sep = ""))Anonymised Tables
# Alan
AnonTitle <- paste(Class, Eval, "Final Grades - Anonymised", sep = " - ")
AnonTable <-
MockTotalsCurvedFinal %>%
subset(select = -c(Student, Teacher)) %>%
arrange(-`Grade (R)`) %>%
knitr::kable(caption = AnonTitle, row.names = F) %>%
row_spec(0, bold = T, color = "white", background = RDFZRed3)%>%
kable_styling(full_width = FALSE,
bootstrap_options = c("striped", "hover", "condensed"),
fixed_thead = TRUE)
save_kable(AnonTable, paste(AnonTitle, ".png", sep = ""))
AnonTable| Anon | Paper 3 | Paper 4 (R) | Total (R) | Grade (R) | Letter |
|---|---|---|---|---|---|
| A18 | 39 | 45 | 87 | 100.0 | A* |
| A14 | 41 | 45 | 87 | 100.0 | A* |
| A9 | 40 | 42 | 87 | 100.0 | A* |
| A29 | 39 | 42 | 86 | 98.9 | A* |
| A5 | 41 | 40 | 86 | 98.9 | A* |
| A27 | 39 | 41 | 85 | 97.7 | A* |
| A20 | 37 | 42 | 84 | 96.6 | A* |
| A6 | 37 | 41 | 83 | 95.4 | A* |
| A3 | 41 | 37 | 83 | 95.4 | A* |
| A8 | 37 | 41 | 83 | 95.4 | A* |
| A17 | 34 | 38 | 77 | 88.5 | A |
| A22 | 36 | 36 | 77 | 88.5 | A |
| A11 | 33 | 37 | 75 | 86.2 | A |
| A15 | 36 | 33 | 74 | 85.1 | A |
| A23 | 31 | 37 | 73 | 83.9 | A |
| A25 | 31 | 37 | 73 | 83.9 | A |
| A1 | 35 | 31 | 71 | 81.6 | A |
| A2 | 35 | 28 | 68 | 78.2 | B |
| A28 | 32 | 30 | 67 | 77.0 | B |
| A13 | 31 | 31 | 67 | 77.0 | B |
| A26 | 26 | 35 | 66 | 75.9 | B |
| A19 | 28 | 33 | 66 | 75.9 | B |
| A21 | 24 | 33 | 62 | 71.3 | B |
| A12 | 25 | 28 | 58 | 66.7 | C |
| A16 | 26 | 27 | 58 | 66.7 | C |
| A24 | 24 | 23 | 52 | 59.8 | C |
| A7 | 23 | 23 | 51 | 58.6 | D |
| A10 | 19 | 23 | 47 | 54.0 | D |
| A4 | 17 | 25 | 47 | 54.0 | D |
By-Question Analysis and Comments
In the sections below, we analyse the A2 Psychology - Cambridge Mock Exam responses on a question by question basis. For each question, we:
- Show the distribution of student grades obtained
- Show the Cambridge Mark Scheme for the question
- Discuss any general comments on student responses, including where students typically went wrong
- Provide two exemplar responses
- One exemplar response chosen from a student who gave a strong response to the question
- One exemplar response provided by the Teacher (you should note that this exemplar response will be at a much higher level than you would be expected to provide in either AS or A2 Psychology)
Paper 3
Question 1-
QNumber <- 1
FocalData <-
Paper3 %>%
subset(Qnum == QT3$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 3 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT3$Number[QNumber], " - ", QT3$Question[QNumber], " [", QT3$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
- 1 mark for a basic explanation
- Personal space is a portable, invisible boundary surrounding us into which others may not trespass
- 2nd mark for detailed explanation
- It regulates how closely we interact with others, moves with us, and expands and contracts according to the situation in which we find ourselves
Exemplar Response (Student)
Personal space is the invisible boundary that people have during social interaction. THis boundary expand and contract according to different social situtation. For example, with a friend, the personal space may be smaller while become bigger when interact with a stranger.
Exemplar Response (Teacher)
Personal space refers to a metaphorical boundary that surrounds us that we use to control how we interact with other people. How much personal space is required varies as a function of the situation we are in and the people we are surrounded with - for most people, for example, the acceptable distance from a close friend is different to the acceptable distance from a stranger.
Question 2-
QNumber <- 2
FocalData <-
Paper3 %>%
subset(Qnum == QT3$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 3 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT3$Number[QNumber], " - ", QT3$Question[QNumber], " [", QT3$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
- 1-2 marks for a basic answer with some understanding of the topic area
- 3-4 marks for a detailed answer with a clear understanding of the topic area
Comments
Note that according to the Cambridge Mark scheme you didn’t technically have to say what the 4 Ps are, and actually some Cambridge Graders would not give you the full 4 marks for merely identifying the 4 Ps
Exemplar Response (Student)
This model illustrates 4 components that the seller need to consuder: product (the color, shape, or physical appearance of the goods; whether its attractive), place (where can consume buy the product; the location of store or the shelf position), price (whether the price is rational; are there any discount) and promotion (creating advertisement that could attract consumers). The update version of 4Ps is 4Cs, including client (what consumers demanding), cost (time to buy), communication (two-way dialogue).
Exemplar Response (Teacher)
McCarthy’s “4Ps marketing mix” refers to a model of advertising that focuses on what the retailer can do to increase the sales of their product. THe first P is the product itself, referring to the physical product or service. The second P is the price, which must be appropriate- comparable to what the market will bear and also priced fairly relative to similar products. The third P is place- the location where the product will be sold and the way that the product is displayed in that location. The final P is promotion - the advertising techniques that will be used to call attention to the product.
Question 3-
QNumber <- 3
FocalData <-
Paper3 %>%
subset(Qnum == QT3$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 3 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT3$Number[QNumber], " - ", QT3$Question[QNumber], " [", QT3$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
- Level 3 Response (5-6 marks)
- Candidates will show a clear understanding of the question and will explain two appropriate weaknesses
- Candidates will provide a good explanation with clear detail
- Candidates will show a clear understanding of the question and will explain two appropriate weaknesses
- Level 2 Response (3-4 marks)
- Candidates will show an understanding of the question and will explain one appropriate weakness in detail or two in less detail
- Candidates will provide a good explanation
- Candidates will show an understanding of the question and will explain one appropriate weakness in detail or two in less detail
- Level 3 Response (5-6 marks)
- Candidates will show a basic understanding of the question and will attempt an explanation of weaknesses. There could be a breif explanation of one weakness
- Candidates will provide a limited explanation
- Candidates will show a basic understanding of the question and will attempt an explanation of weaknesses. There could be a breif explanation of one weakness
- Likely weaknesses will be:
- If you use system 1 or 2 in the incorrect type of situation this will lead to a poor decision being made (e.g. using system 1 to buy a car)
- Reductionist- assumes just one type of thinking is used when making a purchase
- Difficult to assess what type of thinking is being used and difficult for consumers to explain their thought processess and may be unaware of some or all of these thoughts
- Difficult to measure in an objective manner what type of thinking is taking place
- If a measurement is done it could lack validity and/or reliability (this needs to be explained in order to be creditable)
- Deterministic- assumes that if the product is a fast purchase, not much thought will have gone into it and the consumer will quickly decide what to buy. It is possible that a consumer may give a quick purchase quite a bit of thought…
Comments
Cambridge seems to have no idea what they want in the mark scheme here
Is this question about weaknesses of the theory itself? Or about weaknesses of system1/system 2 thinking?
The majority of the mark scheme suggests this is a question about weaknesses of the theory… but Point 1 above suggests otherwise…
Of course, point 1 also suggests that whoever wrote this question has no idea what they are talking about (this is a feature of the theory, not a bug! In fact, this is kind of the whole damn point!)
You should answer this question as being about theory if it comes up again
And also, pay attention to what questions are worth…
Exemplar Response (Student)
First, stystem 1 and system 2 thinking has low level of application since this thinking only introduce the definition and the preference of using system 1. It doesn’t tell us how we can change the habit of using system 1 thinking and how can this thinking be used by the retailers to change the consumer behavior.
Second, this thinking only consider the effect of individual factor which is how individuals brain deal with the informattion. HOwever, it doesn’t discuss the role of enviromental and external factors, like whether the information is presented in different forms. Therefore, it only discuss the process of information in terms of one part and its not comprehensive enough
Exemplar Response (Teacher)
The weaknesses of “thinking fast and slow” tend to come from the fact that it is a descriptive theory. First, this means the theory lacks predictive ability - even if we know everything about a consumer and the situation we are in, nothing about Kahneman’s theory gives us an answer about which system they are ultimately going to use. Kahneman suggests that people will more often default to system 1, but this is not strictly determined. From the perspective of a retailer, a theory of consumer behavior that does not offer concrete predictions is not very useful.
Second, it could be argued that the theory is thus untestable - it posits invisible mental states that the actual consumer is not even aware of - and even suggests that when asked to explain their own behavior they will be incorrect. When we observe a consumer making a particular choice, how can we then evaluate which system was being used? Perhaps even after long, slow consideration (system 2) a consumer arrives at a decision we think more closely matches what would be expected of system 1 thinking .
Question 4-
QNumber <- 4
FocalData <-
Paper3 %>%
subset(Qnum == QT3$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 3 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT3$Number[QNumber], " - ", QT3$Question[QNumber], " [", QT3$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
Cambridge 8 Point Describe Mark Scheme
Comments
Discussions of anchoring were really poor - clearly we need to review that study!
You need to at least mention everything from the section to ensure full marks (“comprehensive”) - this shouldn’t be hard when the question gives you all of the headings
Nothing is credited from outside of the named textbook section
Exemplar Response (Student)
Choice heuristics refers to the mental shortcut people takes when making decisions. It can be seaparted as 2 distinct types: availability and representativeness.
Availability heuristics refers to the situation where people take mental shortcuts when purchasing according to how close or how available the product is. For example, people might prefer to shop online since it is easily available anywhere. On the other hand, the representativeness heuristics refers to the situations where people make purchase decision according to the reputation and well-knownness of the product. Which simply means how representative is the product in its category. For example, when people wants to buy a phone, they will be looking for Apple or Huawei as they are more well known.
Another choice heuristics that the consumers take is investigated by Wansink in his anchoring and purchase quantity decision. He conducted 4 experiments. In the first study, he investigated the use of multiple-unit pricing such as “2 for $3” on sales of product through a field experiment in a supermarket. In the second field experiment, he investigated the effect of limiting purchase quantity. THis could be “Only 4 cans gets 12% discount”. In the third study, he investigate the effectiveness of suggestive selling via slogans. In the 4th study, he investigate the effect of suggestions on internal purchase anchor. For example “snickers- buy 12 for your freezer”. All these experimental condition is compared with a control group with no manipulations. THe result shows that all these methods increase sales
In pre-cognitive decisions, the researchers investigated the link between Brain area and decision making. The study uses fMRI to obtain both functional and structural images of the brain. 8 participants are eliminated due to excessive brain movement. THe participants are required to view the products and is gives $20 to purchase any items. The result shows that NAcc which is the brain area link to reward system is activated when intended to buy the product. The MPFC activation suggest a purchase behavior is being made as it represent deicison making area of the brain. On the other hand, Insula that linked with disgust isactivated when the price is to High and hence purchase decisions are not made
Exemplar Response (Teacher)
Psychologists have discovered that far from being entirely rational actors who carefully weigh up all options to make optimal decisions, humans are imperfect consumers who make use of heuristics- simple rules of thumb that allow them to make quick decisions.
Availability and representativeness are two of these such heuristics that consumers use to make choices. The “availability heuristic” refers to the tendency for consumers to make choices based on the product that is most readily cognitively available to them - either a product they have interacted with recently, seen an advertisement for, or are otherwise reminded of by the situation they find themselves in. For example, when choosing a drink, a consumer is likely to be swayed by the ubiquity of advertisements for Coca-Cola, or perhaps the fact that they can simply see a large Coke sign in a nearby shop window. The “representativeness heuristic” operates similarly, in that it influences choices by changing what comes to mind most easily for the consumer. In the case of representativeness, consumers tend to choose products that they feel are the most prototypical for a category, so when they want a cola its likely that they will choose “Coca-Cola”, because it is seen as the “standard” drink in that category. For this reason, companies spend a lot of money both making sure their products are readily available and also attempting to establish themselves as being representative of a product category.
Researchers in this area have also studied “anchoring heuristics” - how people decide how many of a product they should buy when making a purchasing decision. Wansink et al. argue that people have a low default purchase quantity anchor - normally when we are looking to buy something, we will buy one or two of that product. Wansink et al. note that one way that authors can move our purchase quantity anchor is by offering discounts, but that this is of limited utility - it is not beneficial to a retailer to entice a consumer to buy 10 of a prodct if the unit profit is too low. Because of this, Wansink et al suggested that retailers could instead manipulate purchase quantity anchors through suggestive advertising, and they tested a number of methods through field experiments conducted at a series of supermarkets. In brief, they found that consumers could be enticed to shift their purchase quantity anchor upward (buy more of a product) based on multiple-unit pricing, limited purchase quantities, and suggestive slogans.
The most general feature of heuristics is that they are fast and occur without much conscious processing. Because of this, exploring them directly requires observing the brain. In a study by Knutson et al., researchers used fMRI to scan the brains of participants while they were making pruchasing decisions. They confirmed the presence of distinct neural circuits that operating at this fast, unconscious level, including the nucleus accumbens, insula, and mesial prefrontal cortex. They found that preferred products activated the nucleus accumbens (associated with reward) while deactivating the mesial prefrontal cortex (associated with reasoning/decision making). Relatedly, they found that the insula (associated with disgust) was activated when prices were higher than anticipated. The existence of these early processing circuits supports the automatic and unconscious nature of many of our purchase intentions.
Question 5-
QNumber <- 5
FocalData <-
Paper3 %>%
subset(Qnum == QT3$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 3 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT3$Number[QNumber], " - ", QT3$Question[QNumber], " [", QT3$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
- 1 mark for basic explanation of the term/concept
- Learned helplessness is where are individual feels they have no control over a situation
- **2nd mark for detailed explanation*
- Note: to be considered detailed, the response must included somethign about how the negative experiences caused the sense of helplessness to develop
Comments
The second part of this was the tough bit via the mark scheme - but its asking you to focus on the learned part of learned helplessness
Exemplar Response (Student)
Its one of the explanations of depression. People may experiences negative events in the past, and thus they developed idea that they cannot control over bad circumstances. Thus, they form a pattern of feeling lack of control of everything and this could lead to depression.
Exemplar Response (Teacher)
Learned helplessness is a cognitive explanation of depression that focuses on learning. According to Seligman, people have negative experiences that they are unable to do anything about, and from these experiences learn that they are generally powerless to help themselves. This learned experience then becomes generalized to the point that people become depressed - they have learned to feel like they are powerless and can neither help themselves nor be helped by others.
Question 6-
QNumber <- 6
FocalData <-
Paper3 %>%
subset(Qnum == QT3$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 3 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT3$Number[QNumber], " - ", QT3$Question[QNumber], " [", QT3$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
- 1 mark for each correct point made
- Aim was to investigate the effectiveness of operant conditioning
- Reinforce appropriate behavior in schizophrenic patients
- Set up token economy system in a hospital ward
- Patients were given tokens as reward when they behaved appropriately
- Could be exchanged for luxury items
- Positive and negative symptoms were significantly reduced
- Operant conditioning is an effective means of treating people with chronic schizophrenia
Comments
Pay attention to what questions are worth
I may have been a bit harsh here…
This is not a form of milieu therapy (that was the other level of the IV for comparison!)
Exemplar Response (Student)
Token economy is a physical treatment of depression. Ther ear ethree condition in this study: melieu therapy, normal clinical treatment, and token economy. Milieu therapy aimed to let patient help each other in order to deal with schizophrenia. This is a lab experiment, participants are randomly assigned in each group. TOken economy is operated in a clinic. Participant who act a correct behavior is rewarded a token. This token can be used to exchange for clothes, food, etc. Although the result showed that 3 condition can all reduce symptoms of schizophrenia, token economy produced the most significant reduction.
Exemplar Response (Teacher)
The use of a token economy for the treatment of schizophrenia is a behavior technique that attempts to use operant conditioning to reinforce appropriate behaviors in schizophrenic patients, and thus change their behavior in the long term. In this study, which was conducted in a hospital ward, chronic schizophrenic patients were observed by the hospital staff. When they were observed engaging in certain prosocial behaviors (the opposite of typical schizophrenic behaviors) they were rewarded with tokens which could later be exchanged for luxury items. The researchers found that this method of operant conditioning was effective for treating patients with schizophrenia compared to two different control groups, both in terms of limiting positive and negative symptoms of schizophrenia and in terms of predicting future living outcomes.
Question 7-
QNumber <- 7
FocalData <-
Paper3 %>%
subset(Qnum == QT3$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 3 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT3$Number[QNumber], " - ", QT3$Question[QNumber], " [", QT3$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
- Level 3 (5-6 marks)
- Candidates will show a clear understanding of the question and will include one similarity and one difference
- Candidates will provide a good explanation with clear detail
- Level 2 (3-4 marks)
- Candidates will show an understanding of the question and will include one appropriate similarity in detail or one appropriate difference in detail
- Candidates will provide a good explanation
- Level 1 (1-2 marks)
- Candidates will show a basic understanding of the question and will attempt a similarity and/or difference. This could include both but just as an attempt.
- Candidates will provide a limited explanation
Comments
Your answers can’t be included in the question - i.e. “they are both treatments for depression” or “one is about drugs and one is about cognition”
Explain one, not list 3!
This question is asking you for depth, so give it some depth (and again… pay attention to it being a 6 mark question)
Exemplar Response (Student)
Drug treatment for depression is really common: MAOIs and SSRI are two representative drugs. And the cognitive restructuring treatment is the doctor help patients to restruct their thoughts and beliefs then alleviate the symptoms of depression.
The main difference between drugs and cognitive restructuring treatment is the severity fo the side effects. The drugs are directly act on patients’ neurotransmitters and increase the level of neurotransmitter in their brain. Therefore, the side effect of drugs are serious, like the insomnia, headache, and so on. However, there is no physical side effect of cognitive restructing treatment. Because it use the communication and other cognition way to change patients’ thoughts
The main similarity of these two treatments are high level of application. Both of them are really effective and could help to alleviate depression in some extent. For drugs, it is easy to take and majority of depressed patients treat depression with the help of drugs. And the cognitive restructuring treatment is also really effective and could help to deal with depression. THerefore, these two methods have really high application.
Exemplar Response (Teacher)
Drug treatments for depression focus on manipulating levels of neurotransmitters in the brain (usually functionally increasing the amount of serotonin, e.g. by the use of an SSRI), while cognitive restructure treatment focuses on treating faulty cognitions that lead to depressive symptoms.
One similarity between these two types of teatment is that they both have a “lag time” before treatment typically becomes effective. In the case of drugs, this is because the neurochemistry takes some time to adapt to the presence of the drug and stabilise. In the case of cognitive restructuring, a patient (with the help of their therapist) must first take some time to identify their faulty cognitions, and then to challenge them repeatedly before there is any significant reduction in symptoms.
One difference between the two is patient involvement, which can have significant implications for treatment. In the case of psychiatric drugs, the patient only needs to be involved to the extent that they actually take their drugs - their role in the experience is passive. In the case of cognitive restructuring, on the other hand, the patient is forced to take an active role- working both with their therapist and alone to challenge their negative cognitions and learn to think about things differently. Because of this, cognitive restructuring, like most cognitive approaches, is more effective with patients who are willing to “buy in” to the process and do the required work.
Question 8-
QNumber <- 8
FocalData <-
Paper3 %>%
subset(Qnum == QT3$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 3 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT3$Number[QNumber], " - ", QT3$Question[QNumber], " [", QT3$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
Cambridge 8 Point Describe Mark Scheme
Comments
You must discuss the named issue…
Structure
This study didn’t kill Little Albert!
Exemplar Response (Student)
There are many explanations in regards to the causation of phobia in recent years and all has their strength and weaknesses.
The genetic explanation of phobia suggests that phobia is passed down through gene and research has shown that most blood phobia patients has first degree relatives who also have blood phobias. This could be argued that it takes the nature side of the debate since children carries the phobic gene of their parents. However, the situation can also be explained through the nurture debate. While children grows up they might observe the phobic reacts of their parent or relative such as fainting or screaming, thus their fear could be learned from the parent as a consequence of overexposure.
For the psychoanalytical explanation of phobia, they suggest that phobias is caused by people retain in one of the sexual development stages. This could be considered to be subjected to the negative influences of reductionism since the researchers tries to use simple theory toe xplain complex mechanism. For example, Little Han’s fear for Horse could be a result of Oedipus complex but it can also be argued that it is caused by witnessing a horse accident and observe the death of a horse. Thus, the explanation is not comprehensive.
The cognitive explanation could be argued to have low generalisability as the only participant is little Albert. Thus ,the sample size is too small for the result of the study to be generalised to the public. Moreover, in the experiment, although the parent has give their informed consent, Little Albert has not been deconditioned so the trauma might be persistent. THis is highly unethical and brings negative perspective to the field of psychology. ALso, the use of children is also debatable as Little Alebrt does not understand his right to withdraw and might suffers from the experiment and cause lifelong fear of rats.
In conclusion, although these explanations are valid in some ways, it is still flawed and incomprehensive. Further research is required for the explanation to improve.
Exemplar Response (Teacher)
The explanations of anxiety and phobic behaviors offered in this course have a number of strengths and weaknesses.
First, the degree to which phobias are caused by nature, rather than nurture, is questionable. Although the study of blood and injection phobics provides evidence that phobic patients are more likely to have close relatives that share the same phobia, the evidence that this phobia is genetically inherited is lacking. According to some, this evidence supports the nature side of the nature-nurture debate, suggesting that blood phobias are likely inherited due to their adaptive value. However this evidence tells us nothing of the sort- it is equally (perhaps more) probably that blood phobics display this behavior because of learning - because they have observed their close relatives reacting phobically to certain types of stimuli and have acquired the phobia vicariously. To make any claims based on genetic relatedness we’d require a proper twin study where identical twins were separated at birth - if in that case we found that they shared the same phobias despite being reared apart, we might be able to say we have compelling evidence for the nature side of the debate.
Related to this is the inability of the explanations offered to say much about individual differences. Naturally, with learning being a key explanation of phobic behaviors (see Little Albert), situational factors are important for explaining phobias, but are situational factors alone enough to explain them? Taken to an extreme we might suggest that we cannot generalise these studies of phobias -after all, Little Albert was only one child, so according to some we should not generalise what we learned about phobias from this study. This is probably not true - when we are talking about learned phobias, we are talking about basic psychological mechanisms - learning works the same in a snail as it does in a human (neurons that fire together wire together) and probably anyone subjected to the same treatment as Albert would develop the same phobias. However it does seem likely that the conditioning of phobic responses and anxiety is probably easier in some people than in others - or at least its possible that there are people that do really have some natural predispositions towards dealing poorly with anxiety. The studies discussed in this section however don’t allow us to explore such a possibility.
There are also some serious ethical issues surrounding these explanations. Both the behavioral approach and the psychodynamic approach to studying phobias involved case studies using young children who could not give proper consent to participate in these studies. Little Albert had a powerful phobia induced in him that became increasingly generalised to all fluffy white things, and was never de-conditioned, potentially developing a debilitating lifelong phobia (we do not know what would have happened, as Albert died of unrelated causes at the age of 6). Although Little Hans was not directly interfered with in the same way, his parents shared inadequately anonymised information about their young son that he did not consent to being shared and that might have been embarassing to him later in life.
Paper 4
Question 1-
QNumber <- 1
FocalData <-
Paper4 %>%
subset(Qnum == QT4$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 4 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT4$Number[QNumber], " - ", QT4$Question[QNumber], " [", QT4$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
- 1 mark for basic answer
- A long-term condition that causes feelings of anxiety about a wide range of situations and issues
- 2 marks for elaboration
- Can include both psychological and physical symptoms that vary from person to person
- Feel anxious most days and often struggle to remember the last time they felt relaxed
Comments
Students lost marks here for failing to differentiate from specific phobias - generalised
Exemplar Response (Student)
Generalised anxiety disorder is a kind of phobia disorder .The patients of this disorder may often feel stress and worry about the surrounding environment. Feeling tired, insomnia, and so on.
Exemplar Response (Teacher)
Generalised anxiety disorder refers to a persistent feeling of anxiety that applies to a wide range of situations and issues, differentiating it from more limited types of anxiety (e.g. agorophobia) and from specific phobias (e.g. button phobia). People who experience generalised anxiety disorder feel anxious most of the time, and this is manifest in both psychological and physical symptoms that vary from person to person but often include insomnia, loss of appetite, and other common symptoms.
Question 2-
QNumber <- 2
FocalData <-
Paper4 %>%
subset(Qnum == QT4$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 4 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT4$Number[QNumber], " - ", QT4$Question[QNumber], " [", QT4$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
- Up to 2 marks for generic advantages/disadvantages
- data can be analysed statistically
- Up to 2 marks for links to anxiety
Comments
Be sure to make an explicit link to the question (assessing anxiety)
Exemplar Response (Student)
An advantage is that if the comparison can be drawn easily. For example, using BIPI and GAD-7 to assess the anxiety, we can get ass core to do the screening or determine the severity of the anxiety disorder.
The disadvantage is that we cannot have a detailed data analysis (analyse the qualitative data). Therefore, we cannot know the cause of the anxiety disorder and the patients’ thoughts.
Exemplar Response (Teacher)
One advantage of using quantitative data to assess anxiety is that it makes screening for anxiety disorders easy and objective. A psychometric test like the GAD-7 can be administered to large groups of people easily and the results easily compared to baseline to determine who should seek additional counseling or medical help.
What quantitative approaches to assessing anxiety gain in ease of use, they lose in depth of explanation - because a patient’s anxiety is boiled down a single number, it loses descriptive power that might help us better understand anxiety. In this case qualitative data - the type likely to be obtained via interviews asking open questions - will likely help us understand an individual’s experience of anxiety in a way that will be more productive for treatment.
Question 3-
QNumber <- 3
FocalData <-
Paper4 %>%
subset(Qnum == QT4$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 4 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT4$Number[QNumber], " - ", QT4$Question[QNumber], " [", QT4$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
- Up to 2 marks for identification of main methods
- Observation - naturalistic, covert, participant, structured
- Field experiment- IVs (number of intruders, number of buffers); DVs(qualitative verbal comments, quantitative behaviors)
- 1 mark for elaboration or example specific to the study.
Comments
Don’t waste too much time on 2 point responses!
Exemplar Response (Student)
They used field experiment. The study conducted in New York and found several queues waiting for bus or restaurant servies. Its in participants normal environment and they don’t know the aim of the study.
Exemplar Response (Teacher)
Milgram et al. conducted a field experiment in New York City exploring the way that people reacted to intruders into queues at a variety of public locations. They used covert observations to collect both quantitative and qualitative data about how people reacted to these queue intrusions.
Question 4-
QNumber <- 4
FocalData <-
Paper4 %>%
subset(Qnum == QT4$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 4 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT4$Number[QNumber], " - ", QT4$Question[QNumber], " [", QT4$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
- 1 mark for each
- physical action
- non-verbal objections
Comments
None
Exemplar Response (Student)
There could also be physical actions and non-verbal objections.
Exemplar Response (Teacher)
In addition to verbal objections, people could also react via non-verbal objections (such as gesturing or rolling their eyes) or via physical actions (like physically attempting to remove the intruder from the line).
Question 5-
QNumber <- 5
FocalData <-
Paper4 %>%
subset(Qnum == QT4$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 4 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT4$Number[QNumber], " - ", QT4$Question[QNumber], " [", QT4$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
- 1 mark for each advantage/disadvantage (however detailed) related to the question (max 4)
- **1 mark for conclusion*
Comments
Cambridge mark scheme is stupid here - you need to provide two strengths and two weaknesses
So treat this like an AS evaluate question but without the evaluation!
Exemplar Response (Student)
The strength is that the problems of reductionism are not appeared. Unlike the genetic explanation, the cognitive explanation acknowledge that schizophrenia can be explained by some biological factors, but it can also be explained by cognitive factor. Therefore, this explanation is more comprehensive. Also, it provides a detailed explanation for hallucination, delusion, and negative symptoms about why its abnormality of self monitoring.
However, the weakness is that this explanation lacks the data proof so this explanation is still not that reliable than other. Also, since its about the problem of faulty mental processes, we need to know how the mental distortion contribute to the schizophrenia. However, through this explanation we only know that we attribute some inner speech or self-generated ideas to external sources without understanding what happened inside our mind.
Therefore, I don’t think it a reliable and reasonable explanation compared to the other two that one with the genetic study data proof and one with the proof of using drugs trial and ???.
Exemplar Response (Teacher)
One strength of the cognitive explanation of schizophrenia is that it is more comprehensive and less reductionist than other explanations - it acknowledges both the biological underpinnings of schizophrenic symptoms and the behavioral manifestations thereof. A second strength is that it explains both positive and negative symptoms of schizophrenia as being caused by a single construct- abnormalities of self monitoring - thus the theory is not unnecessarily complex.
The simplicity of the cognitive explanation is, however, also one of its weaknesses. It relies on discussing cognitive states that may be unconscious, and that are certainly difficult to test objectively. Relatedly, it is not clear why these abnormalities would be causative - i.e. it is not clear that this theory offers an explanation or strong avenues for treatment, rather than being simply a cognitive description of a biologial phenomenon.
In sum, although the cognitive explanation of schizophrenia is simple and comprehensive, I do not believe it to be exceptionally convincing or applicable to treatment or management of the disorder.
Question 6-
QNumber <- 6
FocalData <-
Paper4 %>%
subset(Qnum == QT4$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 4 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT4$Number[QNumber], " - ", QT4$Question[QNumber], " [", QT4$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
Cambridge 10 Point Design Mark Scheme
Comments
Some good work here, often falling short of 10 marks
Often no identification of how Impulse Control Disorder was established, or what specific disorder(s) was being studied
Sampling techniques not adequately described
Poor operationalisation of DV
Exemplar Response (Student)
The aim of this lab experiment is to investigate the intensity of positive feelings about stealing in people with and without kleptomania. The IV is people with kleptomania and people without. DV is the intensity of positive feelings which is measure on a “positive feeling scale”(6 items, 1(not at all) -5 (very much) rating scale) 50 participants with kleptomania were drawn from an ICD disorder health center. They were recently diagnose with DSM-5 and K-SAS scale which assess their feelings, thoughts and behaviour related to stealing with points scaled questions from 0-4/5 (0=no symptoms, 4/5 = frequent, enduring symptom). They were selected through random sampling. 40 normal people are selected through opportunity sampling in a shopping mall. They were signed informed consent before begin and were told they can withdraw anytime they like.
This experiment took place in a lab. Before starting, participants do the positive feeling scale to measure the baseline of their feelings. Each individual were separated in each quiet room paired with one researcher. The researcher will take them into muscle relaxation and virtualization. They will visualize being exposed to a supermarket where they are many people around, and they are a rally of shoes in from of them, that maybe ? for shoplifting. Then they imagined themselves stealing the shoes and put them in a bag and go away. This process will be for 10 minutes long. After the visualization, they complete the positive feeling scale again which ask them to rate how intense they feel, including please, excitement and so on. Any side effects are observed by researchers. If participants show sign of side effect like headache, the visualization will be terminated. After this, patients were debriefed and give confidentiality consent to their data and alternate therapy is given – if they suffer. There will 2 independent researchers analyze the data. They will subtract the score from baseline to see the changes, mean changes of score in two condition are calculated and plotted on a bar chart for data comparison.
Exemplar Response (Teacher)
Aim- To investigate the the relationship between positive feelings about an event in people with Impulse Control Disorders
Hypothesis- Participants with impulse control disorders will experience more extreme positive feelings about events overall than participants without impulse control disorders, and specifically the most extreme positive feelings related to their focal disorder (kleptomania)
Independent Variable (Disorder)- Presence or absence of diagnosis of an Impulse Control Disorder (kleptomania), screened via scores on the Kleptomania Assessment Scale
Independent Variable II (Type of Activity)- Questions about either focal (stealing) or non-focal (other, e.g. eating) activity
Dependent Variable (Positive Feelings)- Intensity of positive feelings on a forced-choice scale from 1 (intense negative feeling) to 8 (intense positive feeling) from a self-report interview rating feelings towards a variety of behaviors
Sample- Drawn from two samples: ICD Sample = 20 patients selected via opportunity sampling from an outpatient treatment center for kleptomania, all of whom scored >15 on the Kleptomania Assessment Scale. Normal Sample = 20 age-matched patients selected via opportunity sampling for the community of the Peking University, none of whom scored >5 on the Kleptomania Assessment scale. Participants with active addictions or other psychological disorders were excluded from the experiment.
Design/Method- Independent Measures Design/Natural Experiment
Procedures-
1- Prior to the study, all subjects to screening via Kleptomania Assessment Scale
2- Participants are informed about the nature of the study and asked to give informed consent for their participation
3- Participants are brought into an interview room with 2 researchers and free to move around. After a brief informal chat (to make them comfortable), researchers conduct a structured interview exploring feelings about a variety of behaviors. For each behavior:
- Participant is asked to close their eyes and imagine engaging in the behavior the behavior
- Asked to provide a rating for their feelings on a scale from 1 to 8
- Also asked to describe their feelings (open question) about the event
- Participant is asked to close their eyes and imagine engaging in the behavior the behavior
4- Participants were fully debriefed about their participation in the study and given the opportunity to retroactively withdraw their data/participation
Data Analysis- A linear mixed effects regression will be used to analyse intensity of feelings towards the behavior with Disorder Status (Kleptomania vs. No Kleptomania) as a between-subjects factor, Type of Activity (stealing vs. non-stealing) as a within subjects factor, and Participant as a random factor (with by-participant random intercepts and slopes as per a maximal model).
Probable Results- There was no significant main effect of type of activity - the intensity of feelings towards stealing and non-stealing events was equal overall. However, there was a significant main effect of Disorder Status - participants with Kleptomania reported more intense subjective feelings of pleasure than non-Kleptomaniacal participants. There was also a significant interaction of Disorder Status x Type of Activity - although Kleptomaniacs reported more intense positive feelings overall, this difference was heightened for activities around stealing.
Question 7-
QNumber <- 7
FocalData <-
Paper4 %>%
subset(Qnum == QT4$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 4 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT4$Number[QNumber], " - ", QT4$Question[QNumber], " [", QT4$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
Cambridge 10 Point Design Mark Scheme
Comments
Discussions of methodological evidence are typically stronger here
* Don’t waste too much time talking about “weaknesses” of your design
* Instead talk about “design decisions” - reasons for the choices you have made (including potentially weighing up limitations)
June was pretty lenient on the grading of this question!
Exemplar Response (Student)
Psychological: For the feeling state-theory, it suggest that if we win the gambling when we first time do the gambling, we would generate really positive emotions and the thoughts that “I’m a really lucky and powerful person and I”m going to win the gambling all the time“. Therefore, people’s intensity of positive feeling can be very different if they get the impulse control disorder. Also, in order to be diagnosed with this order, people need to have a clear mood modification and salience or tolerance toward this event. THerefor, using the questionnaire that ask”how excited they are " can be a good question to test for the mood changes.
Methodological: The study uses independent measures design and this can avoid the problem of order effect. Also the study uses likert style questions - its easier to calculate the total score and determine the intensity of positive feelings. The analysis of quantitative data can also be a really objective and easy comparison since we only need to compare the statistical data. Also, the volunteer sampling can avoid the problem of high drop-out rate since participants are willing to take part in it.
Exemplar Response (Teacher)
Psychological: The feeling-state theory suggests that people develop Impulse Control Disorders like kleptomania because they produce powerful feeling-states that they are then driven to pursue in the future. A person who is powerless in their normal life, for example, might experience an overwhelming feeling of power and competency when they successfully steal something. Because this is so different from their normal powerless state, they will be driven to pursue this behavior in the future. One observation of ICDs is that they are often comorbid, which suggests that they are underpinned by common psychological dysfunctions - this experiment tests feeling states around not only focal behaviors, but other enjoyable behaviors to help determine if ICD sufferers are generally more likely to develop intense feeling states that become disorders.
Methodological: Because it is impossible (or at least unethical) to modify the disorder state of experimental participants, this study makes use of an independent measures design in a natural experiment - making use of natural variation in kleptomaniacal behavior. This can introduce some fairly substantial participant variables, which we attempt to get around by using age-matched controls to assure that we are at least comparing somewhat similar samples. We also make use of both qualitative and quantitative data to ensure that statistical analyses are sinmple and our hypothesis can be tested directly, while ensuring that we collect additional open-ended descriptions from our participants that might lead to a better understanding of the phenomenology of ICD, or perhaps point towards directions for further research.
Question 8-
QNumber <- 8
FocalData <-
Paper4 %>%
subset(Qnum == QT4$Number[QNumber]) %>%
mutate(GradeF = factor(Grade, levels = c(0:Value[QNumber])))
FocalTitle <- paste(Class, Eval, "Paper 4 - ", sep = " - ")
FocalSubTitle <- paste("Question ", QT4$Number[QNumber], " - ", QT4$Question[QNumber], " [", QT4$Value[QNumber], " marks]", sep = "")
ggplot(data=FocalData, aes(x=GradeF)) +
geom_bar(stat = "count", position = pd, width = 0.8, fill= RDFZRed3) +
scale_x_discrete(drop = FALSE) +
labs(x="Question Score", y="Count") +
theme_alan() +
ggtitle(wrapper(FocalTitle, width = 50),
subtitle = wrapper(FocalSubTitle, width = 100))
Cambridge Mark Scheme
- Typical CAN responses
- There are many similarities between shopper behaviour in different countries
- The cognitive processes of shoppers are the same, such as choice blindness
- Experiments anywhere can control many extraneous variables
- Typical CANNOT responses
- Specific conditions created in one country cannot be recreated in a different country (e.g. outdoor temperatures, etc.)
- People are different: the reasons for shopping; different priorities, amounts of money, etc.
- Shops themselves are different: large malls, supermarkets
Cambridge 12 Point Essay Mark Scheme
Comments
Lots of you ran out of time… (sorry)
Many of you went into great detail describing Hall or other studies - you’re wasting your time, this isn’t a describe question!
This question isn’t all about Hall - Hall is just one example. Use examples from other field experiments where appropriate and to demonstrate your knowledge of the course materials
Give an introduction and a conclusioN! Use topic statements! Structure!
Exemplar Response (Student)
Consumer psychology is an important part in people’s life and the subject deserves greater attention. Currently, it is controversial that whether the result in one country could be generalised to another country. The question remains debatable.
One one hand, the result of the study in one country cannot be generalised in another country. FOr example, in Hall et al. the research is conducted in Europe or North America Country. Which main has white populations. Thus, the results of the study might not be able to generalise to Asian countries where jams are not the major food choice. People might be blind due to unfamiliarness of the product, thus the study cannot be generalised.
Moreover, demand characteristics are also a major issue in field experiments. In Hall et al. the researchers told the participants that this is a quality control test. THus, the participant might not report that they have noticed the difference to convince the researchers that the product has good quality. THus, the results gathered from the study cannot generalise to the wider population.
On the other hand, researchers can always argue that human shares the same decision making process of system 1 and system 2, which simply means that when we are making daily decisions we tends to use system 1, which is thinking fast while we think slow when making more important decisions. Thus, when people are tasting jam, they all utilises the system 1 thinking process, and make a fast decision on this mundane task. Thus, it could be argued that the result could be generalised to other countries.
Furthermore, field experiments has good ecological validity as the experiment is conducted in the participant’s normal environment. Despite the cultural differences, the supermarket environment in each countries are mainly the same. Thus, it could be argued that all participants experienced the same influence of the environment. Thus, the result of the study could be generalise to the Daily life of participants all over the world.
In conclusion, the statement is not completely correct as in some instances results can be generalise to other countries. However, the statement is still valid in most cases.
Exemplar Response (Teacher)
The degree to which results can be generalised from one sample or one experiment to the larger population from which they are drawn (and indeed, to the target population of “all people”) is a vexing question in all of psychology. However, I believe it to be a relatively more easy question to answer in consumer psychology: although we can probably learn some general principles of consumer behavior that can generalise between populations, we should be incredibly wary of generalising from WEIRD populations to the rest of the world.
First, there are a number of fairly trivial differences that make generalisation difficult. It should go without saying that we could probably not replicate Hall’s choice blindness study in a country that does not eat jam - consumers might not have developed a palate that would make it possible for them to differentiate the tastes of different flavors of jam made from unfamiliar fruits. Similarly, consumers in sub-Saharan Africa might not respond the same to “upscale classical music” in restaurant settings because they might not share the same connotations between certain types of music and certain behaviors. I call these differences trivial because perhaps we could find analogues to test in other societies (e.g. choice blindness of different sesame oils in China) that would reveal the same basic principles at play, however it is worth noting that this is almost never done.
This highlight a more major problem for generalising these kinds of studies - unlike some other parts of psychology, consumer psychology rarely deals with what would be called fundamental mental processes - the kinds of cognition and behavior that appear to be natural and shared by all people (and potentially all organisms). When we study phobias, for example, we understand that these can be caused by learning, and that learning is fundamentally identical in all organisms. Not so for consumer psychology. Even well-worn disctinctions like system 1 and system 2 thinking beggar belief when applied to other cultures - does it even make sense to ask about “trivial purchase decisions” when we’re talking about people who live in a country where every consumer choice is made on the knife’s edge of survival? Does it make sense to ask questions about what color of pen someone would prefer to keep in a society where a pen has no value or where even owning a single pen would be considered a luxury? It could be argued, for example, that system 1 and 2 thinking only develop in the west because of our relative wealth and the ability to make trivial economic decisions, rather than that they unmask some fundamental features of human cognition.
Of course, not all of consumer psychology or its many field experiments fall prety to this trap: explorations of primacy and recency effects in menu positions or other advertisements at least pay lip service to fundamental psychological processes and should be generalisable. Here however, the realities of life in other places often get in the way - there may indeed be room for primacy and recency effects to be used in advertising in parts of the developing world, but without the money to make meaningful decisions these fundamental processes might not be important.
This brings us to a fundamental question at the heart of consumer psychology - is it about understanding the cognition of humans at a fundamental level, or simply about maximising profit and explaining behaviors of the consumer class. One might argue that it does not matter that we cannot generalise teh findings of consumer psychology to people in non-industrialised countries - they aren’t the target of the research anyways. This answer might be satisfying to some, but it is problematic both for retailers (who miss out on expanding markets) and for those interested in psychology, who mistake results obtained from limited studies with tiny samples for profound insight into human cognition.
The biggest problems with generalising results like those of Hall et al. does not come from some fundamental wrong-headedness about the idea of “human cognition”, but simply from lazy science - replicating studies with new populations or materials to determine the degree to which they actually tell us something about psychology, rather than simply jumping to grand conclusions based on limited data. Certainly there are cultural differences that will fail to be captured with WEIRD populations, but until we extend our findings both within and outside of WEIRD populations, we cannot actually be certain how large these differences are, and that jeopardizes the certainty of any and all conclusions we can reach in consumer psychology
Comments
No major comments, just be sure to earn your 2!