research design. variables independent variable dependent variable levels responses
Post on 20-Dec-2015
242 views
TRANSCRIPT
Research design
Variables
• Independent variable
• Dependent variable
• Levels
• Responses
Variables and conditions
Subjects are given two types of constructions and are asked to decide whether the given sentence is grammatical:
(1) a. I gave him it. Construction 1b. I gave her the book.c. …
(2) a. I gave to him it. Construction 2b. I gave to her the note you sent me.c. …
Variables and conditions
IV (two conditions) DV (forced choice task)
Construction 1Construction 2
a. grammaticalb. ungrammatical
Variables and conditions
Subjects are asked to complete copular sentences with a relative clause. The predicate nominals of the copular clauses belong to three different semantic types: (1) animate/human (2) inanimate/object (3) place.
(1) a. This is the man __ b. This is the ball __c. This is the place __
Variables and conditions
IV DV
1. This is the man __ 2. This is the ball __3. This is the place __
a. SUBJ relative clauseb. DO relative clausec. IO relative claused. OBL relative clausee. GEN relative clause
Variables and conditions
IV DV
1. This is the man __ 2. This is the thing __3. This is the place __
a. SUBJ relative clauseb. DO relative clausec. IO relative claused. OBL relative clausee. GEN relative clause1. I saw the man __
2. I saw the thing __3. I saw the place __
Interaction
Condition 1 Condition 2
SUBJ 3.5DO 3.2IO 2.7OBL 2.2GEN 0.6
SUBJ 2.5DO 3.8IO 3.2OBL 0.5GEN 0.5
Types of data
• Nominal/categorical data
• Ordinal data
• Interval data
Type of analysis
• Correlational analysis
• Difference test
Type of analysis
Correlational test Difference test
Pearson‘s rKendall‘s tau
T-testANOVA
Confound variable
(1) … the man who talked to Mary.
(2) … the car that caused the accident.
(3) … the man who Mary talked to.
(4) … the car that Peter bought.
Confound variable
(1) … the man who talked to Mary.
(2) … the car that caused the accident.
(3) … the man who Mary talked to.
(4) … the car that Peter bought.
Confound variables
• Control
• Randomization
Sampling
• Simple random sampling
• Stratified random sampling
• Systematic sampling
• Cluster sampling
Related and independent design
• Within subjects design – related design – repeated measures design
• Between subjects design – unrelated design – independent design
Advantages of within subjects design
• Reduction of inter-individual differences
• Fewer subjects
Disdvantages of within subjects design
• Subjects recognize the purpose of the study.
• Subjects get tired, frustrated, excited.
• Subjects get habituated to the task.
Responses to IVs/conditions can influence each
other:
Counterbalancing
• ABBA
• AB - BA
Counterbalancing serves to eliminate the
ordering effect.
Counterbalancing
1.ABC
2.ACB
3.BAC
4.CAB
5.BCA
6.CBA
Experimental design
A child language researcher wants to find out if the
meaning of the head of a relative clause influences the
interpretation of the acquisition of relative clauses in
early child speech. Specifically, he wants to know if
animate and inanimate head nouns affect children’s
interpretation of relative clauses. In this study, he
concentrates on the two most frequent types of relative
clauses in which the subject and object are relativized
(i.e. expressed by the relative pronoun).
Experimental design
(1) Das ist der Mann, der das Mädchen gestern gesehen hat.
(2) Das ist der Mann, den das Mädchen gestern gesehen hat.
(3) Das ist der Ball, der das Mädchen am Kopf getroffen hat.
(4) Das ist der Ball, den das Mädchen mit dem Kopf getroffen hat.
Experimental design
(1) Das ist der Mann, der das Mädchen gestern gesehen hat.
(2) Das ist der Mann, den das Mädchen gestern gesehen hat.
(3) Das ist der Ball, der das Mädchen am Kopf getroffen hat.
(4) Das ist der Ball, den das Mädchen mit dem Kopf getroffen hat.
Experimental design
Animate head Inanimate head
Subject
Object
Central tendency
Data: 2, 3, 3, 3, 4, 6, 6, 9, 12, 13, 13
Mean: 2+3+3+3+4+6+6+9+12+13+13
11= 6.72
Median: 2, 3, 3, 3, 4, 6, 6, 9, 12, 13, 13 = 6
Mode: 2, 3, 3, 3, 4, 6, 6, 9, 12, 13, 13 = 3
Variance and SD
(x1 – x)2
N- 1
Variance
S words
12345678
3749129114
Variance
S words
12345678
3749129114
59 / 8 = 7.4 (mean)
Variance
S words (=X1 – Xmean)
12345678
3749129114
3 – 7.4 7 – 7.44 – 7.49 – 7.412 – 7.49 – 7.411 – 7.44 – 7.4
59 / 8 = 7.4 (mean)
Variance
S words (=X1 – Xmean) d1
12345678
3749129114
3 – 7.4 7 – 7.44 – 7.49 – 7.412 – 7.49 – 7.411 – 7.44 – 7.4
–4.4–0.4–3.41.64.61.63.6–3.4
59 / 8 = 7.4 (mean)
Variance
S words (=X1 – Xmean) d1
12345678
3749129114
3 – 7.4 7 – 7.44 – 7.49 – 7.412 – 7.49 – 7.411 – 7.44 – 7.4
–4.4–0.4–3.41.64.61.63.6–3.4
59 / 8 = 7.4 (mean)
0 / 8 = 0
Variance
S words (=X1 – Xmean) d1 d12 (residuals)
12345678
3749129114
3 – 7.4 7 – 7.44 – 7.49 – 7.412 – 7.49 – 7.411 – 7.44 – 7.4
–4.4–0.4–3.41.64.61.63.6–3.4
19.360.1611.562.5621.162.5612.9611.56
59 / 8 = 7.4 (mean)
0 / 8 = 0 81.87
Variance
81.87
8 - 1= 11.7
Standard Deviation
81.87
8 - 1= 3.42
Standard Deviation
70% of all data points fall within 1 SD.
Mean +/- SD = range of 70% of the data
7.4 +/- 3.42 = 3.98 – 10.82 words