1
00:00:00,560 --> 00:00:05,160
Speaker 1: Welcome back, everybody. Well, we did assessments this week. It

2
00:00:05,240 --> 00:00:07,960
took a look a lot of statistics, So this week's

3
00:00:08,320 --> 00:00:11,240
ten question practice quiz will consist of some of the

4
00:00:11,240 --> 00:00:14,679
material we covered earlier in the week. Question number one

5
00:00:14,720 --> 00:00:18,679
A researchers investigating in rare neurological syndrome and tracks three

6
00:00:18,760 --> 00:00:23,399
affected individuals over a five year period, collecting detailed qualitative

7
00:00:23,399 --> 00:00:28,320
behavioral neuroimaging data. What is the prominent primary methodological limitation

8
00:00:28,440 --> 00:00:35,079
of this research design A low external validity B higher

9
00:00:35,159 --> 00:00:44,880
risk of carryover effects or see low internal consistency.

10
00:00:43,200 --> 00:00:45,479
Speaker 2: A is the correct answer case.

11
00:00:45,479 --> 00:00:49,119
Speaker 1: Studies involving a very rich, small, specific, simple size offer

12
00:00:49,240 --> 00:00:53,000
rich detail, but heavily limit the ability to generalize the

13
00:00:53,079 --> 00:00:54,960
findings and the broader populations.

14
00:00:55,039 --> 00:00:56,320
Speaker 2: That's external validity.

15
00:00:59,600 --> 00:01:02,880
Speaker 1: Aganizational psychologists evaluates a new workplace.

16
00:01:02,920 --> 00:01:06,799
Speaker 2: This is question number two wellness and let me repeat that.

17
00:01:06,840 --> 00:01:07,680
So question number two.

18
00:01:07,680 --> 00:01:12,560
Speaker 1: An organizational psychologist evaluates a new workplace wellness initiative by

19
00:01:12,560 --> 00:01:17,599
measuring employee job satisfaction scores monthly for six months before

20
00:01:17,640 --> 00:01:20,959
the program starts and monthly for six months following the

21
00:01:21,000 --> 00:01:25,200
completion of the program. This research structure is best classified

22
00:01:25,200 --> 00:01:31,439
as what a regression does continuity design, B factorial design

23
00:01:32,079 --> 00:01:34,640
or C time series design.

24
00:01:38,159 --> 00:01:41,519
Speaker 2: The answer is C. Time series design involves.

25
00:01:41,159 --> 00:01:45,719
Speaker 1: Tracking trends over time by taking multiple distinct observations both

26
00:01:45,760 --> 00:01:51,239
before and after a specific intervention is introduced. So it's

27
00:01:51,280 --> 00:01:55,200
a time zeries design, again, involving tracking trends over time

28
00:01:55,359 --> 00:01:59,879
by taking multiple distinct observations before and after a specific

29
00:02:00,040 --> 00:02:06,959
intervention is introduced. Question number three. A clinical trial evaluates

30
00:02:06,959 --> 00:02:11,520
the impact of treatment types CBT versus medication and client

31
00:02:11,599 --> 00:02:16,639
age group adolescent versus adult on depression severity scores. The

32
00:02:16,680 --> 00:02:21,840
study utilizes a two times two factorial design. Which of

33
00:02:21,840 --> 00:02:26,719
the following statements is true regarding this design, A it

34
00:02:26,759 --> 00:02:33,199
evaluates two independent variables and each containing two levels. It

35
00:02:33,240 --> 00:02:40,319
evaluates four independent variables containing one level. Or C it

36
00:02:40,360 --> 00:02:46,639
evaluates a single independent variable across four distinct measurement periods.

37
00:02:50,319 --> 00:02:54,840
If you said A, you're right, it's A evaluates two

38
00:02:54,960 --> 00:02:58,719
independent variables containing two levels. See in factorial notation, the

39
00:02:58,800 --> 00:03:03,280
number of digits represents the number the number of independent variables,

40
00:03:03,319 --> 00:03:07,400
so we started with two slash times two factorial design.

41
00:03:09,000 --> 00:03:12,439
Two represents the digital the number of independent variables and

42
00:03:12,560 --> 00:03:15,120
the value of Each digit indicates how many levels that

43
00:03:15,159 --> 00:03:19,599
specific variable has. Each has two, so two times two.

44
00:03:19,639 --> 00:03:25,439
Factorial design is two iv and two levels. Question number four.

45
00:03:26,199 --> 00:03:33,080
A developmental psychologist wants to completely minimize error variance caused

46
00:03:33,080 --> 00:03:38,960
by individual participant differences like personality or intelligence while testing

47
00:03:39,000 --> 00:03:44,840
the cognitive effects of three different educational mobile apps. So again,

48
00:03:45,560 --> 00:03:49,520
demental psychologists wants to completely minimize error variants caused by

49
00:03:49,520 --> 00:03:55,159
individual differences while testing the cognitive effects of three different

50
00:03:55,280 --> 00:03:56,520
educational mobile apps.

51
00:03:56,520 --> 00:03:59,240
Speaker 2: Which design choice is most appropriate.

52
00:04:00,560 --> 00:04:04,919
Speaker 1: Between subjects design with random assignment, A cross sectional design

53
00:04:05,039 --> 00:04:11,639
utilizing stratified sampling, A within subjects design utilizing counterbalancing.

54
00:04:16,399 --> 00:04:18,439
Speaker 2: If you said C, that's correct.

55
00:04:18,639 --> 00:04:22,759
Speaker 1: So within subjects design, the same participants experience all conditions,

56
00:04:23,439 --> 00:04:26,199
meaning each person acts as their own baseline control and

57
00:04:26,240 --> 00:04:31,600
individual differences are perfectly controlled. Counterbalancing must be used alongside

58
00:04:31,639 --> 00:04:38,720
it to eliminate the risk of order effects. Let me

59
00:04:38,759 --> 00:04:40,480
see if this might help clear it up. A little

60
00:04:40,480 --> 00:04:43,560
bit of statistics is in your game. Look at you

61
00:04:43,639 --> 00:04:47,319
mix the experimental setup for this. Imagine you're testing the

62
00:04:47,319 --> 00:04:51,519
impact of three different desktop environments on a participant's ability

63
00:04:51,560 --> 00:04:57,079
to spot errors in a complex data set. Condition as

64
00:04:57,079 --> 00:05:00,959
a standard clutter desktop with constant pop up on your computer.

65
00:05:01,480 --> 00:05:06,399
Condition P is a clean minimalist desktop with all notification silence,

66
00:05:06,480 --> 00:05:11,279
and Condition C is a specific ambient desktop featuring a subtle,

67
00:05:11,560 --> 00:05:13,240
loopable background.

68
00:05:12,759 --> 00:05:14,879
Speaker 2: Video of a fireplace.

69
00:05:16,240 --> 00:05:18,920
Speaker 1: So again you're looking to see how different digital environments

70
00:05:19,000 --> 00:05:20,399
affect a person's focus.

71
00:05:21,319 --> 00:05:22,680
Speaker 2: Instead of recruiting.

72
00:05:22,240 --> 00:05:25,199
Speaker 1: Three separate groups, to recruit one group of thirty participants

73
00:05:25,199 --> 00:05:27,399
and every single one of them completes the task under

74
00:05:27,480 --> 00:05:31,120
all three conditions, so you have phase one acting as

75
00:05:31,160 --> 00:05:34,959
their own baseline. Because participant five might naturally have the

76
00:05:34,959 --> 00:05:37,920
attention span of a zen master, where Participant twelve gets

77
00:05:37,920 --> 00:05:42,560
distracted by shiny objects. Between subjects design would require massive

78
00:05:42,560 --> 00:05:46,480
sample sizes to average out these personal traits, but with

79
00:05:46,560 --> 00:05:49,959
a within subject design, Participant five's performance in the clutter

80
00:05:50,079 --> 00:05:52,839
environment is compared directly to their own performance in the

81
00:05:52,839 --> 00:05:58,399
minimalist environment, So individual differences in baseline focus, intelligence, or

82
00:05:58,439 --> 00:06:01,560
even how much coffee they drank are controlled because the

83
00:06:01,560 --> 00:06:06,160
person remains identical across all three conditions. I hopefully that

84
00:06:06,199 --> 00:06:10,040
helps that one number five. A study finds a statistically

85
00:06:10,079 --> 00:06:16,680
significant correlation P value of less than point zero one

86
00:06:17,360 --> 00:06:19,600
and R value of point four five.

87
00:06:19,920 --> 00:06:21,439
Speaker 2: That's the correlation number.

88
00:06:22,279 --> 00:06:25,399
Speaker 1: Between a clinician's total years of experience in patient program

89
00:06:25,439 --> 00:06:26,240
retention rates.

90
00:06:26,240 --> 00:06:27,000
Speaker 2: That's what they find.

91
00:06:27,040 --> 00:06:30,560
Speaker 1: So a study found statistically significant correlation between a clinician's

92
00:06:30,600 --> 00:06:35,639
total years of experience and patient program retention rates. A

93
00:06:35,720 --> 00:06:38,800
critic points out that highly experienced clinicians also tend to

94
00:06:38,839 --> 00:06:41,839
practice and well through areas with better public transit, which

95
00:06:41,879 --> 00:06:44,959
could explain the retention. We would call this, by the way,

96
00:06:44,959 --> 00:06:51,319
a confounding variable. This critique highlights which methodological issue A

97
00:06:51,759 --> 00:06:57,480
A third variable problem, B low statistical conclusion validity or

98
00:06:57,600 --> 00:07:01,720
d I mean sorry C order effect.

99
00:07:03,000 --> 00:07:03,639
Speaker 2: The answer is A.

100
00:07:03,839 --> 00:07:06,720
Speaker 1: The third variable problem occurs when an unmeasured variable, in

101
00:07:06,759 --> 00:07:10,720
this case, the neighborhood, wealthy, and transit, correlates with both

102
00:07:10,720 --> 00:07:13,879
the predictory and the outcomes, creating a misleading causal appearance

103
00:07:13,920 --> 00:07:18,319
and providing why correlation does not equal causation.

104
00:07:19,439 --> 00:07:19,959
Speaker 2: Number six.

105
00:07:20,040 --> 00:07:23,839
Speaker 1: The researcher measures treatment outcomes using a classification system where

106
00:07:23,879 --> 00:07:27,839
symptoms severity is marked as mild, moderate, or severe. What

107
00:07:28,000 --> 00:07:32,879
scale of measurement does this classification system represent? A nominal,

108
00:07:33,800 --> 00:07:41,360
b ordinal, or C interval. If you said ordinal, you're right.

109
00:07:41,600 --> 00:07:46,000
Ordinal scales categorize data into a distinct rank order. However,

110
00:07:46,120 --> 00:07:49,639
the exact mathematical distance between the ranks is only unequal

111
00:07:49,759 --> 00:07:54,439
and unknown. Number seven, an investigator wants to determine if

112
00:07:54,439 --> 00:07:58,720
a new therapeutic module causes a reduction in generalized anxiety.

113
00:07:59,279 --> 00:08:03,240
Due to a minister limitations at the clinic, the researcher

114
00:08:03,279 --> 00:08:06,000
cannot randomly assign patients to conditions.

115
00:08:06,279 --> 00:08:07,879
Speaker 2: Instead, they utilize.

116
00:08:07,480 --> 00:08:11,240
Speaker 1: A pre existing therapy group for the treatment and a

117
00:08:11,360 --> 00:08:16,879
different pre existing group as the control. The study is

118
00:08:16,879 --> 00:08:21,600
best described as what a true experimental design A or

119
00:08:21,720 --> 00:08:27,040
B quasi experimental non equivalent control group design or see

120
00:08:27,600 --> 00:08:32,639
a naturalistic observation study.

121
00:08:32,919 --> 00:08:34,080
Speaker 2: If you said B, you're right.

122
00:08:34,159 --> 00:08:36,639
Speaker 1: This design compares a treatment and control group, but lacks

123
00:08:36,639 --> 00:08:40,440
the random assignment component, relying instead on pre existing intact groups,

124
00:08:40,840 --> 00:08:45,120
hence quasi experimental non equivalent control group design.

125
00:08:46,559 --> 00:08:47,159
Speaker 2: Number eight.

126
00:08:47,320 --> 00:08:51,480
Speaker 1: A study conducted in an ultra controlled lab environment demonstrates

127
00:08:51,480 --> 00:08:56,159
a massive reduction in depressive symptoms. However, when clinicians attempt

128
00:08:56,159 --> 00:08:59,639
to implement the same protocol in busy, fast paced community

129
00:08:59,639 --> 00:09:03,600
mental health centers, the treatment fails to show significant benefits.

130
00:09:03,759 --> 00:09:08,120
This discrepancy indicates that the original lab study possessed what

131
00:09:11,120 --> 00:09:16,399
high internal validity but low external validity, low internal validity

132
00:09:16,399 --> 00:09:21,639
but high external validity, or see low statistical conclusion validity

133
00:09:21,639 --> 00:09:23,519
and high attrition bias.

134
00:09:26,360 --> 00:09:27,679
Speaker 2: If you said a you're right.

135
00:09:27,799 --> 00:09:32,279
Speaker 1: Highly controlled lab settings isolate variables well to establish clear causation,

136
00:09:33,240 --> 00:09:36,639
high internal validity, but often failed to generalize into more

137
00:09:36,679 --> 00:09:43,279
complicated real world conditions low external validity. Question number nine.

138
00:09:43,440 --> 00:09:45,960
When interpreting the results of a two way a nova

139
00:09:46,559 --> 00:09:49,960
within a factorial design, a researcher notes, of the combination

140
00:09:50,080 --> 00:09:54,399
of treatment type and client age produces a distinct, non

141
00:09:54,559 --> 00:10:01,080
parallel pattern of treatment outcomes on the graph. Demonstrates what

142
00:10:01,320 --> 00:10:06,080
the presence of a significant what a main effect for

143
00:10:06,120 --> 00:10:11,879
treatment type be interaction effect or ce selection threat. So,

144
00:10:11,919 --> 00:10:14,039
when interpreting the results of a two way in nova

145
00:10:14,080 --> 00:10:17,039
within a factorial design, a researcher notes, of the combination

146
00:10:17,120 --> 00:10:20,879
of treatment type and client age produces a distinct, non

147
00:10:20,919 --> 00:10:25,759
parallel pattern of treatment outcomes on the graph. If you

148
00:10:25,799 --> 00:10:28,360
said interaction effect, You're right. It occurs when the impact

149
00:10:28,360 --> 00:10:31,159
of one independent variable depends directly on the level of

150
00:10:31,200 --> 00:10:35,799
another in the pendent variable. Visually, a significant interaction is

151
00:10:35,840 --> 00:10:40,320
always represented by non parallel or intersecting lines on a graph,

152
00:10:40,600 --> 00:10:45,559
and lastly to number ten to ensure that carryover effects,

153
00:10:46,840 --> 00:10:51,399
practice effects, or participant fatigue do not distort the data.

154
00:10:51,480 --> 00:10:55,240
In a within subject study evaluating three separate memory strategies,

155
00:10:55,919 --> 00:10:59,000
the investigator alters the sequence in which the strategies are

156
00:10:59,039 --> 00:11:03,320
practiced across different subgroups. This control technique is known as what.

157
00:11:05,039 --> 00:11:07,320
Speaker 2: So again, we're looking at? What is it called?

158
00:11:08,320 --> 00:11:13,080
Speaker 1: When the research you're evaluating within subjects three separate memory strategies,

159
00:11:13,120 --> 00:11:16,519
the investigator alters a sequence in which the strategies are practiced.

160
00:11:16,919 --> 00:11:24,279
Is it random sampling, B stratification or C counterbalancing? If

161
00:11:24,279 --> 00:11:29,320
you said, see counterbalancing your rights systematically varies. Counterbalancing systematically

162
00:11:29,440 --> 00:11:33,399
varies or randomizes the presentation order of conditions across participants

163
00:11:33,399 --> 00:11:36,639
to distribute and cancel out potential order effects like fatigue

164
00:11:36,720 --> 00:11:42,240
or practice effects, because those are going to have an impact. Well,

165
00:11:42,360 --> 00:11:43,919
that's it for your ten practice questions.

166
00:11:43,919 --> 00:11:45,039
Speaker 2: I hope you did well.

167
00:11:45,519 --> 00:11:48,159
Speaker 1: If not, you can always hear it again and continue

168
00:11:48,159 --> 00:11:48,720
to practice.

169
00:11:49,440 --> 00:11:50,559
Speaker 2: Thanks everyone for listening.